Monday, November 30, 2009

Mapping crime statistics UK style

Just to complement my latest post on Australian mapshups with crime statistics, a quick look on how the topic is dealt with in the UK – on the official map of crimes and antisocial behaviour from National Policing Improvement Agency. The map was launched with a great fanfare by Home Office Minister on 20th October, 2009.

The application shows monthly and quarterly crime statistics for the last couple of years (by type) for policing regions in England and Wales. Crime levels in neighbourhoods, within the policing regions, are shown on a Bing Map as shaded thematic overlays. Users can compare statistics of selected area to other locations as well as download the data as csv files.

Due to high demand the site went down frequently within the first 24 hrs after the launch but now performance is quite reasonable. However, the developers did not put much effort into generating neighbourhood boundaries since they do not align well with each other. As well, there may be some underlying issues with the data collection methodology – my UK based colleague pointed out that an area in his neighborhood with no population at all (ie. local park) is shown on the map as having the highest crime rate. This is of some concern since it my potentially influence purchase decisions and the demand for properties in the area. What is traditionally seen as a very desirable feature (ie. parkland) is now indicated as a hive of criminal activities. It highlights the perils of using statistics indiscriminately.

I believe that Australian developers participating in the MashupAustralia tackled technical and presentation challenges for equivalent set of crime statistics pretty well, and to a very high standard.

Australia’s love for private space

A news item making rounds in the media this morning is that Australians have the largest houses in the entire world (statistically speaking). A research commissioned by CommSec and conduced by the Australian Bureau of Statistics concluded that the average floor area of new dwellings in Australia hit a record 214.6 square metres in the last financial year. In the US the average size of a residential dwelling is only 201.5 square metres which still compared favourably to Denmark which has Europe’s biggest dwellings with an average floor area of 137sq m, Greece (126sq m) or the Netherlands (115.5sq m). Homes in the UK are the smallest in Europe at 76sq m.

Australia's largest dwellings were built in Victoria with an average 224.5sq m floor area, followed by Western Australia, Queensland, Northern Territory, NSW, Tasmania, South Australia and the ACT. However, the biggest houses are in NSW where the average new house built in 2008-09 was 262.9sq m, and in Queensland (253sq m). We love the space but does it makes us happier? Just think about the cleaning…

Sunday, November 29, 2009

MashupAustralia highlights Pt 4

One dataset that generated quite a lot of interest amongst developers was New South Wales Crimes by offence type, month and Local Government Area (1995-2008). It is quite a comprehensive and complex bit of information and developers took a range of different approaches in presenting that information in their applications. All in all, there were 7 applications built with crime information as the main theme and several more where crime statistics were used as a complementary information. Here is a review of the most interesting submissions.

NSW Crime Explorer presents statistics in tables, graphs and on Google Map as thematic overlays (only 2008 data). Tables are very well laid out and are very easy to read despite that they contain very comprehensive set of information. I particularly like little graphs in each table row depicting trends over the 10 year period. There are also links to line graphs with monthly stats for each crime type (generated with Google Visualisation tools). A drop down list of all Local Government Areas enables easy navigation between locations of interest. The map is very basic but a selection of overlays for each crime type clearly shows where the offences were committed in 2008. The author has chosen 15 categories for data classification and it slightly blurs the clarity as to which areas are crime hotspots. Overall a very comprehensive presentation.

CrimeFinder is a mashup created with Silverlight and Bing Map. Thematic map shades Local Government Areas according to crime rate or absolute numbers of crimes committed. Different crime types can be selected from a drop-down list. Mouse-click on a particular area brings up a line graph showing comparative trends in crimes for NSW and the selected LGA. Mouse-over function highlights particular boundary and brings up summary information for the area. Slider filters allow setting start and end dates for data to be presented on the map as well as enable adjusting transparency level of the overlays. Redraw of LGA boundaries on the map is very smooth on zooming and panning which is quite an achievement considering it is all vector data.

NSW Crime Map demonstrates yet another approach to presenting crime statistics with Google Map and Google Visualisation tools. The number of crimes committed in major NSW regions, in any given month, is represented as dots on the map. Size of those dots is proportional to the number of committed offences. Different crime types are selectable from a list next to the map and selection of dates is done with a slider tool. Dynamic graph under the map displays monthly counts of incidents between 1995 to 2008.

How Safe Is Your Suburb is an application created with commercial software integrated with Google Map and Flash graphics. Crime data is presented in four different ways: based on geographic distribution (thematic map with data table), as a cross tabulation of offence by type and year (pie chart and data table), as a cross tabulation of year of the offence and type by Local Government Area (with line graph and data tables) and as summary statistics highlighting the most dangerous regions in NSW.

NSW Crime is a very simple mashup presenting crime data in two columns and a Google Map with location points depicting user selected Local Government Areas. The first data column contains a list of LGAs and counts of all crimes committed in those areas. The second column lists various types of offences and counts for a selected LGA. There is also an option to specify time range for calculating statistics. Simple approach yet allowing to show a wealth of information.

Saturday, November 28, 2009

Malocclusion: Disease of Civilization, Part VIII

Three Case Studies in Occlusion

In this post, I'll review three cultures with different degrees of malocclusion over time, and try to explain how the factors I've discussed may have played a role.

The Xavante of Simoes Lopes

In 1966, Dr. Jerry D. Niswander published a paper titled "The Oral Status of the Xavantes of Simoes Lopes", describing the dental health and occlusion of 166 Brazilian hunter-gatherers from the Xavante tribe (free full text). This tribe was living predominantly according to tradition, although they had begun trading with the post at Simoes Lopes for some foods. They made little effort to clean their teeth. They were mostly but not entirely free of dental cavities:
Approximately 33% of the Xavantes at Simoes Lopes were caries free. Neel et al. (1964) noted almost complete absence of dental caries in the Xavante village at Sao Domingos. The difference in the two villages may at least in part be accounted for by the fact that, for some five years, the Simoes Lopes Xavante have had access to sugar cane, whereas none was grown at Sao Domingos. It would appear that, although these Xavantes still enjoy relative freedom from dental caries, this advantage is disappearing after only six years of permanent contact with a post of the Indian Protective Service.
The most striking thing about these data is the occlusion of the Xavante. 95 percent had ideal occlusion. The remaining 5 percent had nothing more than a mild crowding of the incisors (front teeth). Niswander didn't observe a single case of underbite or overbite. This would have been truly exceptional in an industrial population. Niswander continues:
Characteristically, the Xavante adults exhibited broad dental arches, almost perfectly aligned teeth, end-to-end bite, and extensive dental attrition. At 18-20 years of age, the teeth were so worn as to almost totally obliterate the cusp patterns, leaving flat chewing surfaces.
The Xavante were clearly hard on their teeth, and their predominantly hunter-gatherer lifestyle demanded it. They practiced a bit of "rudimentary agriculture" of corn, beans and squash, which would sustain them for a short period of the year devoted to ceremonies. Dr. James V. Neel describes their diet (free full text):
Despite a rudimentary agriculture, the Xavante depend very heavily on the wild products which they gather. They eat numerous varieties of roots in large quantities, which provide a nourishing, if starchy, diet. These roots are available all year but are particularly important in the Xavante diet from April to June in the first half of the dry season when there are no more fruits. The maize harvest does not last long and is usually saved for a period of ceremonies. Until the second harvest of beans and pumpkins, the Xavante subsist largely on roots and palmito (Chamacrops sp.), their year-round staples.

From late August until mid-February, there are also plenty of nuts and fruits available. The earliest and most important in their diet is the carob or ceretona (Ceretona sp.), sometimes known as St. John's bread. Later come the fruits of the buriti palm (Mauritia sp.) and the piqui (Caryocar sp.). These are the basis of the food supply throughout the rainy season. Other fruits, such as mangoes, genipapo (Genipa americana), and a number of still unidentified varieties are also available.

The casual observer could easily be misled into thinking that the Xavante "live on meat." Certainly they talk a great deal about meat, which is the most highly esteemed food among them, in some respects the only commodity which they really consider "food" at all... They do not eat meat every day and may go without meat for several days at a stretch, but the gathered products of the region are always available for consumption in the community.

Recently, the Xavante have begun to eat large quantities of fish.
The Xavante are an example of humans living an ancestral lifestyle, and their occlusion shows it. They have the best occlusion of any living population I've encountered so far. Here's why I think that's the case:
  • A nutrient-rich, whole foods diet, presumably including organs.
  • On-demand breast feeding for two or more years.
  • No bottle-feeding or modern pacifiers.
  • Tough foods on a regular basis.
I don't have any information on how the Xavante have changed over time, but Niswander did present data on another nearby (and genetically similar) tribe called the Bakairi that had been using a substantial amount of modern foods for some time. The Bakairi, living right next to the Xavante but eating modern foods from the trading post, had 9 times more malocclusion and nearly 10 times more cavities than the Xavante. Here's what Niswander had to say:
Severe abrasion was not apparent among the Bakairi, and the dental arches did not appear as broad and massive as in the Xavantes. Dental caries and malocclusion were strikingly more prevalent; and, although not recorded systematically, the Bakairi also showed considerably more periodontal disease. If it can be assumed that the Bakairi once enjoyed a freedom from dental disease and malocclusion equal to that now exhibited by the Xavantes, the available data suggest that the changes in occlusal patterns as well as caries and periodontal disease have been too rapid to be accounted for by an hypothesis involving relaxed [genetic] selection.
The Masai of Kenya

The Masai are traditionally a pastoral people who live almost exclusively from their cattle. In 1945, and again in 1952, Dr. J. Schwartz examined the teeth of 408 and 273 Masai, respectively (#1 free full text; #2 ref). In the first study, he found that 8 percent of Masai showed some form of malocclusion, while in the second study, only 0.4 percent of Masai were maloccluded. Although we don't know what his precise criteria were for diagnosing malocclusion, these are still very low numbers.

In both studies, 4 percent of Masai had cavities. Between the two studies, Schwartz found 67 cavities in 21,792 teeth, or 0.3 percent of teeth affected. This is almost exactly what Dr. Weston Price found when he visited them in 1935. From Nutrition and Physical Degeneration, page 138:
In the Masai tribe, a study of 2,516 teeth in eighty-eight individuals distributed through several widely separated manyatas showed only four individuals with caries. These had a total of ten carious teeth, or only 0.4 per cent of the teeth attacked by tooth decay.
Dr. Schwartz describes their diet:
The principal food of the Masai is milk, meat and blood, the latter obtained by bleeding their cattle... The Masai have ample means with which to get maize meal and fresh vegetables but these foodstuffs are known only to those who work in town. It is impossible to induce a Masai to plant their own maize or vegetables near their huts.
This is essentially the same description Price gave during his visit. The Masai were not hunter-gatherers, but their traditional lifestyle was close enough to allow good occlusion. Here's why I think the Masai had good occlusion:
  • A nutrient-dense diet rich in protein and fat-soluble vitamins from pastured dairy.
  • On-demand breast feeding for two or more years.
  • No bottle feeding or modern pacifiers.
The one factor they lack is tough food. Their diet, composed mainly of milk and blood, is predominantly liquid. Although I think food toughness is a factor, this shows that good occlusion is not entirely dependent on tough food.

Sadly, the lifestyle and occlusion of the Masai has changed in the intervening decades. A paper from 1992 described their modern diet:
The main articles of diet were white maize, [presumably heavily sweetened] tea, milk, [white] rice, and beans. Traditional items were rarely eaten... Milk... was not mentioned by 30% of mothers.
A paper from 1993 described the occlusion of 235 young Masai attending rural and peri-urban schools. Nearly all showed some degree of malocclusion, with open bite alone affecting 18 percent.

Rural Caucasians in Kentucky

It's always difficult to find examples of Caucasian populations living traditional lifestyles, because most Caucasian populations adopted the industrial lifestyle long ago. That's why I was grateful to find a study by Dr. Robert S. Corruccini, published in 1981, titled "Occlusal Variation in a Rural Kentucky Community" (ref).

This study examined a group of isolated Caucasians living in the Mammoth Cave region of Kentucky, USA. Corruccini arrived during a time of transition between traditional and modern foodways. He describes the traditional lifestyle as follows:
Much of the traditional way of life of these people (all white) has been maintained, but two major changes have been the movement of industry and mechanized farming into the area in the last 25 years. Traditionally, tobacco (the only cash crop), gardens, and orchards were grown by each family. Apples, pears, cherries, plums, peaches, potatoes, corn, green beans, peas, squash, peppers, cucumbers, and onions were grown for consumption, and fruits and nuts, grapes, and teas were gathered by individuals. In the diet of these people, dried pork and fried [presumably in lard], thick-crust cornbread (which were important winter staples) provided consistently stressful chewing. Hunting is still very common in the area.
Although it isn't mentioned in the paper, this group, like nearly all traditionally-living populations, probably did not waste the organs or bones of the animals it ate. Altogether, it appears to be an excellent and varied diet, based on whole foods, and containing all the elements necessary for good occlusion and overall health.

The older generation of this population has the best occlusion of any Caucasian population I've ever seen, rivaling some hunter-gatherer groups. This shows that Caucasians are not genetically doomed to malocclusion. The younger generation, living on more modern foods, shows very poor occlusion, among the worst I've seen. They also show narrowed arches, a characteristic feature of deteriorating occlusion. One generation is all it takes. Corruccini found that a higher malocclusion score was associated with softer, more industrial foods.

Here are the reasons I believe this group of Caucasians in Kentucky had good occlusion:
  • A nutrient-rich, whole foods diet, presumably including organs.
  • Prolonged breast feeding.
  • No bottle-feeding or modern pacifiers.
  • Tough foods on a regular basis.
Common Ground

I hope you can see that populations with excellent teeth do certain things in common, and that straying from those principles puts the next generation at a high risk of malocclusion. Malocclusion is a serious problem that has major implications for health, well-being and finances. In the next post, I'll give a simplified summary of everything I've covered in this series. Then it's back to our regularly scheduled programming.

Tuesday, November 24, 2009

Malocclusion: Disease of Civilization, Part VII

Jaw Development During Adolescence

Beginning at about age 11, the skull undergoes a growth spurt. This corresponds roughly with the growth spurt in the rest of the body, with the precise timing depending on gender and other factors. Growth continues until about age 17, when the last skull sutures cease growing and slowly fuse. One of these sutures runs along the center of the maxillary arch (the arch in the upper jaw), and contributes to the widening of the upper arch*:

This growth process involves MGP and osteocalcin, both vitamin K-dependent proteins. At the end of adolescence, the jaws have reached their final size and shape, and should be large enough to accommodate all teeth without crowding. This includes the third molars, or wisdom teeth, which will erupt shortly after this period.

Reduced Food Toughness Correlates with Malocclusion in Humans

When Dr. Robert Corruccini published his seminal paper in 1984 documenting rapid changes in occlusion in cultures around the world adopting modern foodways and lifestyles (see this post), he presented the theory that occlusion is influenced by chewing stress. In other words, the jaws require good exercise on a regular basis during growth to develop normal-sized bones and muscles. Although Dr. Corruccini wasn't the first to come up with the idea, he has probably done more than anyone else to advance it over the years.

Dr. Corruccini's paper is based on years of research in transitioning cultures, much of which he conducted personally. In 1981, he published a study of a rural Kentucky community in the process of adopting the modern diet and lifestyle. Their traditional diet was predominantly dried pork, cornbread fried in lard, game meat and home-grown fruit, vegetables and nuts. The older generation, raised on traditional foods, had much better occlusion than the younger generation, which had transitioned to softer and less nutritious modern foods. Dr. Corruccini found that food toughness correlated with proper occlusion in this population.

In another study published in 1985, Dr. Corruccini studied rural and urban Bengali youths. After collecting a variety of diet and socioeconomic information, he found that food toughness was the single best predictor of occlusion. Individuals who ate the toughest food had the best teeth. The second strongest association was a history of thumb sucking, which was associated with a higher prevalence of malocclusion**. Interestingly, twice as many urban youths had a history of thumb sucking as rural youths.

Not only do hunter-gatherers eat tough foods on a regular basis, they also often use their jaws as tools. For example, the anthropologist and arctic explorer Vilhjalmur Stefansson described how the Inuit chewed their leather boots and jackets nearly every day to soften them or prepare them for sewing. This is reflected in the extreme tooth wear of traditional Inuit and other hunter-gatherers.

Soft Food Causes Malocclusion in Animals

Now we have a bunch of associations that may or may not represent a cause-effect relationship. However, Dr. Corruccini and others have shown in a variety of animal models that soft food can produce malocclusion, independent of nutrition.

The first study was conducted in 1951. Investigators fed rats typical dry chow pellets, or the same pellets that had been crushed and softened in water. Rats fed the softened food during growth developed narrow arches and small mandibles (lower jaws) relative to rats fed dry pellets.

Other research groups have since repeated the findings in rodents, pigs and several species of primates (squirrel monkeys, baboons, and macaques). Animals typically developed narrow arches, a central aspect of malocclusion in modern humans. Some of the primates fed soft foods showed other malocclusions highly reminiscent of modern humans as well, such as crowded incisors and impacted third molars. These traits are exceptionally rare in wild primates.

One criticism of these studies is that they used extremely soft foods that are softer than the typical modern diet. This is how science works: you go for the extreme effects first. Then, if you see something, you refine your experiments. One of the most refined experiments I've seen so far was published by Dr. Daniel E. Leiberman of Harvard's anthropology department. They used the rock hyrax, an animal with a skull that bears some similarities to the human skull***.

Instead of feeding the animals hard food vs. mush, they fed them raw and dried food vs. cooked. This is closer to the situation in humans, where food is soft but still has some consistency. Hyrax fed cooked food showed a mild jaw underdevelopment reminiscent of modern humans. The underdeveloped areas were precisely those that received less strain during chewing.

Implications and Practical Considerations

Besides the direct implications for the developing jaws and face, I think this also suggests that physical stress may influence the development of other parts of the skeleton. Hunter-gatherers generally have thicker bones, larger joints, and more consistently well-developed shoulders and hips than modern humans. Physical stress is part of the human evolutionary template, and is probably critical for the normal development of the skeleton.

I think it's likely that food consistency influences occlusion in humans. In my opinion, it's a good idea to regularly include tough foods in a child's diet as soon as she is able to chew them properly and safely. This probably means waiting at least until the deciduous (baby) molars have erupted fully. Jerky, raw vegetables and fruit, tough cuts of meat, nuts, dry sausages, dried fruit, chicken bones and roasted corn are a few things that should stress the muscles and bones of the jaws and face enough to encourage normal development.

* These data represent many years of measurements collected by Dr. Arne Bjork, who used metallic implants in the maxilla to make precise measurements of arch growth over time in Danish youths. The graph is reproduced from the book A Synopsis of Craniofacial Growth, by Dr. Don M. Ranly. Data come from Dr. Bjork's findings published in the book Postnatal Growth and Development of the Maxillary Complex. You can see some of Dr. Bjork's data in the paper "Sutural Growth of the Upper Face Studied by the Implant Method" (free full text).

** I don't know if this was statistically significant at p less than 0.05. Dr. Corruccini uses a cutoff point of p less than 0.01 throughout the paper. He's a tough guy when it comes to statistics!

*** Retrognathic.

Secondguessing Murdoch

I follow any information on Mr Murdoch’s attempts to introduce charging for news with great interest. Mostly as a passive observer but occasionally I put forward my thoughts as a comment to an article. There is a lot of speculation on what his strategy will be but hints are getting stronger. Yesterday Guardian (UK) published a piece quoting Mr Murdoch's son James, chief executive of News Corp Europe and Asia, saying that “…the industry will see a shift from merely selling journalism direct to end-consumers - or readers - to a market in which news organisations 'wholesale' their news to other organisations who can use it for their own ends.”

It is a very interesting comment from my perspective since it supports my own conclusions as to Mr Murdoch probable strategy. I commented on this in response to an article published in The Australian on 19 October in which the author reported that News Limited is working on top secret projects and suggested the company is planning to build new websites and associated services based on social network principles. My comment was never published ("censorship" of a kind?) but here it is what I submitted:

Interesting “insider” perspective however, I fail to see how these “new product concepts” you refer to will differ from myriad of online websites that make money from other sources than advertising. Nothing was stopping The News to develop these special purpose or niche products for the last decade or so. Big opportunities are gone and are milked by competition (eg. real estate and other online classifieds). Value adding basic (and free) news does not give The News distinctive advantage – anyone can do it.

With due respect, I don’t think what you describe reflects Mr Murdoch’s true intentions as it would be really “more of the same” strategy for The News. It is unlikely that he had “new and exciting sites” in mind when he had spoken these words: "…The aggregators and plagiarists will soon have to pay a price for the co-opting of our content.” Taken literary, it implies wholesale cross media news/content “feed” subscription model rather than merely “sites … built in the style of social networks”.

The latest round of news is that Mr Murdoch and Microsoft are discussing potential deal which would see the technology group pay for exclusive rights to content from News Limited stable of newspapers to attract visitors to its Bing search engine. The deal would exclude Google and other search engines from indexing any content from Mr Murdoch’s sites. Neither Microsoft nor News Corp are commenting on their talks.

Google to index data from mashups

Last week Google announced its plans to log information from mashups that are using JavaScript Maps API v2. The information will then be used within the main Google Map site. Basic premise is similar to “My Maps” where user generated content can be searched and displayed on a map along other options (eg. locations, businesses or real estate).

Details are scarce but it appears Google will index only location and content of the markers and/or infowindows that are displayed on hundreds of thousands of mashups. There is an option to opt out from this arrangement and Google has indicated it will apply the following rules:

  1. We only index data from maps that have been viewed by many unique users. For example, maps only viewed by you and a few friends won't be indexed.
  2. If your page is protected by robots.txt, we will not index your content.
  3. You can opt-out of the logging by specifying "indexing=false" when loading the API.
  4. If you are a Maps API Premier customer, we will not index data from your maps, unless you opt into indexing by passing in "indexing=true".

The exact day the new arrangement will take effect will be announced in due course.

First spotted on Google Maps Mania

Monday, November 23, 2009

Free data - sign of times…

It’s official. Starting from April 2010, UK citizens and the rest of the world will be able to openly access maps from Ordnance Survey as well as plethora of interpretive geographical data such as crime, health and education statistics by postcode. After many years of significant revenue from a successful commercial model of licensing government geographical information to value added resellers, UK government has decided to change its approach and make the information available online for free.

Ordnance Survey monopoly on GIS data in UK will end although some may argue that its position was already heavily undermined by the success of OpenStreetMap project – a community driven initiative to provide free high resolution vector data in competition with OS. Smaller players and website developers will be the winners as this opens up new opportunities for mashing up the information into myriads of specialised online applications.

Ordnance Survey is a £116 million a year enterprise and now part of this revenue will be forgone for “a wider good”. And the burden of maintaining high quality geographic information will have to be shifted to the UK Government (ultimately taxpayers) as the activity will no longer be funded by end users. However, the argument is that the overall commercial benefit to UK economy will be greater than the lost revenue stream.

There is similar attempt to liberate government data in Australia with Government 2.0 Taskforce initiative. It is not a first such attempt - Spatial Data Access and Pricing Policy from 2001 is still in place. It allowed free access to quite a range of geographic information in the past but it is rather very difficult to assess its economic benefits.

It is quite obvious from past experience that just releasing the data will not lead to any tangible benefits. There must be a framework put in place at the same time for managing and improving that data (ie. either big money from the government or big crowdsourcing initiatives, as in case of OpenStreetMap project). Otherwise there is a danger that we will end up in a big mess… with everyone maintaining their own sets of data (hence multiplying the efforts) and creating own set of problems (one only need to look as issues with postcode data in Australia to understand what it can lead to... ).

Tuesday, November 17, 2009

Malocclusion: Disease of Civilization, Part VI

Early Postnatal Face and Jaw Development

The face and jaws change more from birth to age four than at any other period of development after birth. At birth, infants have no teeth and their skull bones have not yet fused, allowing rapid growth. This period has a strong influence on the development of the jaws and face. The majority of malocclusions are established by the end this stage of development. Birth is the point at which the infant begins using its jaws and facial musculature in earnest.

The development of the jaws and face is very plastic, particularly during this period. Genes do not determine the absolute size or shape of any body structure. Genes carry the blueprint for all structures, and influence their size and shape, but structures develop relative to one another and in response to the forces applied to them during growth. This is how orthodontists can change tooth alignment and occlusion by applying force to the teeth and jaws.

Influences on Early Postnatal Face and Jaw Development

In 1987, Miriam H. Labbok and colleagues published a subset of the results of the National Health Interview survey (now called NHANES) in the American Journal of Preventive Medicine. Their article was provocatively titled "Does Breast-feeding Protect Against Malocclusion"? The study examined the occlusion of nearly 10,000 children, and interviewed the parents to determine the duration of breast feeding. Here's what they found:

The longer the infants were breastfed, the lower their likelihood of major malocclusion. The longest category was "greater than 12 months", in which the prevalence of malocclusion was less than half that of infants who were breastfed for three months or less. Hunter-gatherers and other non-industrial populations typically breastfeed for 2-4 years, but this is rare in affluent nations. Only two percent of the mothers in this study breastfed for longer than one year.

The prevalence and duration of breastfeeding have increased dramatically in the US since the 1970s, with the prevalence doubling between 1970 and 1980 (NHANES). The prevalence of malocclusion in the US has decreased somewhat in the last half-century, but is still very common (NHANES).

Several, but not all studies have found that infants who were breastfed have a smaller risk of malocclusion later in life (1, 2, 3). However, what has been more consistent is the association between non-nutritive sucking and malocclusion. Non-nutritive sucking (NNS) is when a child sucks on an object without getting calories out of it. This includes pacifier sucking, which is strongly associated with malocclusion*, and finger sucking, which is also associated to a lesser degree.

The longer a child engages in NNS, the higher his or her risk of malocclusion. The following graph is based on data from a study of nearly 700 children in Iowa (free full text). It charts the prevalence of three types of malocclusion (anterior open bite, posterior crossbite and excessive overjet) broken down by the duration of the NNS habit:

As you can see, there's a massive association. Children who sucked pacifiers or their fingers for more than four years had a 71 percent chance of having one of these three specific types of malocclusion, compared with 14 percent of children who sucked for less than a year. The association between NNS and malocclusion appeared after two years of NNS. Other studies have come to similar conclusions, including a 2006 literature review (1, 2, 3).

Bottle feeding, as opposed to direct breast feeding, is also associated with a higher risk of malocclusion (1, 2). One of the most important functions of breast feeding may be to displace NNS and bottle feeding. Hunter-gatherers and non-industrial cultures breast fed their children on demand, typically for 2-4 years, in addition to giving them solid food.

In my opinion, it's likely that NNS beyond two years of age, and bottle feeding to a lesser extent, cause a large proportion of the malocclusions in modern societies. Pacifier use seems to be particularly problematic, and finger sucking to a lesser degree.

How Do Breastfeeding, Bottle Feeding and NNS Affect Occlusion?

Since jaw development is influenced by the forces applied to them, it makes sense that the type of feeding during this period could have a major impact on occlusion. Children who have a prolonged pacifier habit are at high risk for open bite, a type of malocclusion in which the incisors don't come together when the jaws are closed. You can see a picture here. The teeth and jaws mold to the shape of the pacifier over time. This is because the growth patterns of bones respond to the forces that are applied to them. I suspect this is true for other parts of the skeleton as well.

Any force applied to the jaws that does not approximate the natural forces of breastfeeding or chewing and swallowing food, will put a child at risk of malocclusion during this period of his or her life. This includes NNS and bottle feeding. Pacifier sucking, finger sucking and bottle feeding promote patterns of muscular activity that result in weak jaw muscles and abnormal development of bony structures, whereas breastfeeding, chewing and swallowing strengthen jaw muscles and promote normal development (review article). This makes sense, because our species evolved in an environment where the breast and solid foods were the predominant objects that entered a child's mouth.

What Can We do About it?

In an ideal world (ideal for occlusion), mothers would breast feed on demand for 2-4 years, and introduce solid food about halfway through the first year, as our species has done since the beginning of time. For better or worse, we live in a different world than our ancestors, so this strategy will be difficult or impossible for many people. Are there any alternatives?

Parents like bottle feeding because it's convenient. Milk can be prepared in advance, the mother doesn't have to be present, feeding takes less time, and the parents can see exactly how much milk the child has consumed. One alternative to bottle feeding that's just as convenient is cup feeding. Cup feeding, as opposed to bottle feeding, promotes natural swallowing motions, which are important for correct development. The only study I found that examined the effect of cup feeding on occlusion found that cup-fed children developed fewer malocclusion and breathing problems than bottle-fed children.

Cup feeding has a long history of use. Several studies have found it to be safe and effective. It appears to be a good alternative to bottle feeding, that should not require any more time or effort.

What about pacifiers? Parents know that pacifiers make babies easier to manage, so they will be reluctant to give them up. Certain pacifier designs may be more detrimental than others. I came across the abstract of a study evaluating an "orthodontic pacifier" called the Dentistar, made by Novatex. The frequency of malocclusion was much lower in children who did not use a pacifier or used the Dentistar, than in those who used a more conventional pacifier. This study was funded by Novatex, but was conducted at Heinrich Heine University in Dusseldorf, Germany**. The pacifier has a spoon-like shape that allows normal tongue movement and exerts minimal pressure on the incisors. There may be other brands with a similar design.

The ideal is to avoid bottle feeding and pacifiers entirely. However, cup feeding and orthodontic pacifiers appear to be acceptable alternatives that minimize the risk of malocclusion during this critical developmental window.

* Particularly anterior open bite and posterior crossbite.

** I have no connection whatsoever to this company. I think the results of the trial are probably valid, but should be replicated.

NearMap Goes Live!

When I was writing my recent post about Stuart Nixon and his latest project: NearMap, the site was not yet operational. But today, to my great excitement, I got an anonymous tip that NearMap is up and running! I could not resist and took it immediately for a test drive.

My first impression is that NearMap looks… so familiar! Creators went to a great length to ensure NearMap has the "look and feel" of other online maps, like Google or Bing. It is quite appropriate because potential users will be immediately familiar with how it works… Yet, there is also plenty of unique features. One such feature is a sliding bar across the top of the map which allows scrolling through imagery acquired at different points in time. Perth has the best selection. Currently high resolution coverage is limited to major Australian cities (Adelaide, Brisbane, Sydney, Melbourne and Perth) but there is enough there to fully demonstrate the quality of imagery and functionality of NearMap. There are a few Landsat mosaics and Blue Marble monthly mosaic series from 2004 to cover the rest of the world for now.

Another unique feature is "Terrain" viewing mode which depicts terrain model derived from high resolution imagery (including buildings!). And when you are in PhotoMap viewing mode and zoom close enough, "Multiview" option is activated which allows observing the object on the ground from 4 different directions. This feature is similar to Bing's Birds-eye view mode.

NearMap street overlay comes from OpenStreetMap. It is worth noting that, unlike other suppliers, NearMap actually encourages users to utilise its imagery to improve the quality of this community generated and maintained vector dataset. There is "Edit" button on the map that takes the user directly to OpenStreetMap site. NearMap can be embedded in other websites with iframe HTML element (as in this blog for example) and specific locations can be bookmarked, shared and/or linked to with a unique URL address.

I am very impressed with NearMap. The level of detail on imagery is amazing and directional views, although not perfectly stitched, are very realistic indeed. You can literary "peek over the fence" and see what your neighbor is up to! I just hope it won't become a big issue that will necessitate degrading of the resolution of publicly viewable imagery. The application is very responsive and there is definitely a great potential for NearMap to carve a big niche in the online mapping market, in direct competition with Google and Bing maps.

I will share with you how to view NearMap images in true 3D but for that you will have to come back next time. :-)

Google enters mobile GPS Navigation Market

So far Google played only on the fringes of the mobile GPS Navigation market with flat Google maps for mobile, integrated with driving directions and live traffic information. But the situation is about to change with the release of Google Map Navigation (beta). This new application comes with everything you'd expect to find in a fully featured GPS navigation system: maps in 3D perspective, turn-by-turn voice guidance and automatic routing. But there is more. Google took advantage of a wide range of products it is already supplying to the public and packaged them together into a potentially very attractive product.

In particular, users will always get the latest maps and there is no need for bulky downloads since maps come directly from Google servers. The application supports simple plain English search, voice search and search for POI along the route. There is live traffic data (an indicator light in the corner of the screen alerts current traffic conditions along your route) as well as high resolution satellite imagery and street view options that show exactly what to expect at the destination.

There is an expectation that Google will release API for Map Navigation to allow developers to build myriad of customised applications and help forthcoming Android phones better compete with the Apple iPhone. Google Map Navigation if free and user only need to cover Internet connection fees to use it. For now, Google Map Navigation can only run on Android 2.0 enabled phones which are very limited in supply. In fact, there is only one handset on the market - Droid from Verizon - on sale since 6th November and only in the U.S.

Originally spotted on: Google Maps Mania

Friday, November 13, 2009

There is money in pictures!

One Australian that is consistently punting on imagery is Stuart Nixon, of ER Mapper fame. He started in 1989 with the invention of the image compression algorithm and creation of software for processing imagery on low cost personal computers. He then followed on with the Image Web Server to enable efficient, and probably still unsurpassed, capability to serve large image files over the Internet to web browsers. The company and all its technology is now in hands of ERDAS Group, after Stuart sold it for undisclosed sum in 2007, but his love affair with pixels has not finished there.

Soon after parting with his old venture Stuart started a new company: NearMap. It developed automated process for fast creation of very high resolution photomaps from aerial imagery (patents pending). The ultimate goal for the company is to cover 700 cities across the World with photomaps updated at least on a monthly basis.

Looks like Stuart Nixon is on the money again with his new idea. NearMap is just about to launch a new service, in competition with Google and other online map providers. Meantime, he sold the company mid last year to Perth based Ipernica for “a job” plus $4 million in cash, plus $12 million worth of Ipernica shares (60 million at a deemed issue price of $0.20 per share) and 12.5 million Ipernica options at a strike price of $0.40, with a two year expiry period. Even with today’s price of Ipernica at $0.12 it is still cool $10 million return. Not bad, not bad at all…

Pixel worth 1000 bits

It took rather long time for raster imagery (maps but also aerial and satellite photography) to gain wide acceptance for use in online GIS applications. Raster imagery used to be just a backdrop to “real data” (ie. vectors) as substantial file size (often in obscure proprietary format) and generally low resolution made it difficult to incorporate imagery into applications and rather impractical to use for more than just the initial “splashback”. The fact that remote sensing, aerial surveys and GIS were traditionally seen as totally separate disciplines was not helping in driving amalgamation of various formats into a single data management and online visualisation solution.

It would be fair to say that the release of Google online mapping technology was a catalyst to initiate change. These days raster imagery is dominating in online mapping applications, especially in those with high traffic such as Google Maps, Bing Maps, Yahoo Maps. The ability to show ground features in detail increased significantly with the advances in image capture technologies, reducing the reliance on vectors to depict such information. The ingenuity of using small image tiles, to overcome the issues with raster data file size and to improve efficiency of online distribution, made it much easier to use raster rather than vector data format for presenting information.

[exmple of raster map: Topo 250K by Geoscience Australia]

Just to clarify, in the traditional online GIS applications both raster images and vectors are presented on a computer screen as images (eg. gif, jpg, png). However, the difference is in how those images are created. In case of vector data (and some raster data with pixel attribute information), all images are generated from data on the fly by a map server (caching techniques are now available for static information to reduce the load) and there is a “link” maintained between what is presented on the image and vector data so if user “clicks” on the line representing eg. road, attribute information specific to that segment of the line can be returned as a query result (eg. name, width, type). In case of raster imagery without attribute information for each pixel, the image is pre-generated in advance (eg. raster topographic map). With this approach, dynamic referencing to attribute information in the source dataset is lost.

[example of map image generated from vector data: Topo 250K by Geoscience Australia]

It is technically possible to use a mix of vector data and raster imagery in a single application to get the best of both approaches (ie. high resolution imagery or nice pre-designed cartographic representation of vector data, delivered fast as tiled images yet still referenced to attribute information in the original dataset) but I have not seen this yet implemented as a general practice. Here is another idea for Google – add “get feature” service to your maps! It would work exactly as “reverse geocoding” service but rather than returning address information for the point (based on cadastre vector dataset) it could also return information on other map features (like roads, parks, buildings, other POI, etc). Creating a “link” to source vector data could also open up the opportunities for all sorts of spatial query service options: “distance”, “area”, but also more complex like “select within”, “adjacent” etc.

With the exception of a handful of technologies, specifically developed for true vector and imagery data streaming, the overwhelming majority of online mapping applications is not capable of processing efficiently vector data into complex images - on the fly and in volumes required for today’s Internet – without a massive hardware infrastructure behind it.

Currently, reliance on true vector data in browsers is limited to presenting a small number of points, polygons or 3D objects, or only to highly specialised applications. There is support for true vector data in a browser environment via Java, Flash or Silverlight but to make it work efficiently it requires sophisticated solutions to handle on-the-fly vector generalisation and local data caching (as mentioned above, only a handful of companies managed to do it, and they are not the industry leaders! Although, I should mentioned that I am very impressed with an online application I have seen recently, developed in Silverlight and showing nicely quite a volume of vector data - I will have to investigate in more detail!

Applications such as Google Map make use of browser’s support for vector data in VML/SVG format but overall, browser processing capabilities are very limited. Therefore, although Google Map will accepts vector data in KML format but if the file is too big (100KB +) Google will convert it to image tiles to speed rendering of the information in the browser. It is appropriate for presentation of static data but will not work with dynamic information (eg. thematic maps) because once vectors are converted to images on initial map load they cannot be changed with a script. If the same amount of data is imported into Google Map application and rendered as true vectors (eg. with GPolygon() function), loading and drawing of the information on the map is rather slow.

There is a new concept emerging for handling of spatial information, regardless whether the source is raster imagery or vector data. It is the concept of a spatial grid. Traditionally, girds were used with Digital Elevation Models (DEM) data. Later they were also found to be applicable in the field of spatial analysis. Now grid concept can also be applied to referencing myriad of attributes to a specific location (cell) on the surface of the Earth – making all the data hierarchical and analysis-ready. If I understand correctly, such organisations as USGS are now planning to start releasing information as referenced grid cells rather than traditional imagery although there are still some challenges in defining those grids, indexing the data, and of course in storage capabilities for high resolution datasets.

The theory and technologies developed for handling imagery will find use in implementation of this new approach. After all, image pixels are a form of a grid. Graphic formats offer more efficient storage and compression capabilities than traditional spatial databases and emergence of graphical processing capabilities (GPU) offer a great hope for very fast analytical capabilities – a new and exciting era in spatial data management, processing and delivery!

MashupAustralia highlights Pt 3

It looks that this will be quite a long series… The competition has just been closed with the total of 78 entries (jumped from 40 just yesterday) - overwhelming majority map based applications. And in the end only one major GIS company decided to participate which is quite a disappointment. It's not the little guys and galls that will make a difference. It is the big end of town that has the capacity and resources to manage the data on ongoing basis and that can bring technology and know-how to enable opening up of government volts of information. In the end, there was probably not enough "fresh carrot" in this contest to bring them aboard. So, let's review those institutional mushups first. Yesterday I finished with entries relating to bushfire information so I will continue with this theme.

Firelocator: map based application showing latest fire related information (satellite hotspots, Country Fire Authority and Rural Fire Services RSS feeds), some basic population stats and photos from Flickr. - built with Silverlight and Bing Map, and surely full GIS at the back end. The application was submitted by Pitney Bowes (formerly MapInfo Australia). Firelocator has a few nice features (like interactive filters for hotspots - based on "confidence" value - or for fire incidents based on type and status). However, there are also a few limitations which is rather disappointing considering that Firelocator carries "TechAmerica Innovator Award" logo. In particular, administrative boundaries overlay is not projected correctly - the further you zoom out the greater the discrepancy. Also, I never liked Bing's onmouseover auto pop-up feature but in case of this application it is frankly quite annoying as it makes map panning very difficult if the map is crowded with makers. Firelocator didn't work for me in FireFox at all (unlike the other Silverlight application I reviewed yesterday).

AuScope Portal: another institutional entry featuring information on Australia's mineral resources, mines and processing centres - built with Google Map and using OGC web services (WMS and WFS). Feels like fully featured online GIS but I found it a bit cumbersome to use (I had real difficulties in displaying and removing layers and especially managing order of layers for display). Great collection of geology related information though!

LocateMe: application submitted by Western Australia Landgate crew. It presents demographic profiles for areas of interest and lists government services available in the area. It uses Google Map as a background layer but is built with OpenLayers - it certainly does not behave like a real Google Map. Once again, the application is styled as a fully featured online GIS with traditional navigation tools - which makes it rather awkward to use (ie. no double click to zoom, no zoom slider tool, requires selection of buttons to either zoom in or out or pan). It could have been quite an interesting application if it wasn't for so many layers being restricted with a password. And I couldn't get demographic profiles to load at all.

EasyData: new website released by the Department of Trade and Economic Development of South Australia that provides information on sixty social, environmental and economic indicators, and allows regional comparisons. It has a basic Flash map to enable users to navigate to area of interest and lots of graphs of statistical and economic information. It may be a bit too heavy for lower end computers though...

Data Aggregator: very slick and well built Google Map application that allows displaying almost all datasets released for MashupAustrlia competition, and more. Easy to navigate and certainly much easier to use than any of the above mentioned mapping mushups. Which is quite surprising since the application is using hierarchical and expandable layer selector, very similar to the other applications mentioned today. A unique feature of Data Aggregator is a couple of custom controls for changing display options for markers and vector overlays.

To be continued…

Location specific time

Dynamically updated data, like eg. weather information on site or various RSS feeds, are published with a time stamp. It is customary that time is expressed as a Greenwich Mean Time (GMT), a universal reference so anyone, in any part of the world, can determine exactly "when this thing happened". Very handy concept but in order to make the most of the information, some recalculations are often required. There are sophisticated server side scripts to deal with time but JavaScript also has a few handy functions to help with the task.

Before I move on to examples, just a quick reference to one more important concept: epoch. It is defined as time kept internally by a computer system, expressed as the number of time units that have elapsed since a specified date. Unix time epoch, used by UNIX, Linux, other UNIX-like systems, Mac OS X, as well as Java, JavaScript, C and PHP Programming languages, started on January 1, 1970 and is usually expressed in milliseconds. The epoch is what makes time related calculations so simple…

You can find many example of various JavaScript time functions by simply “googling” the term however, there is not much practical information on how to apply them. Here are a few transformations I found very useful:

Current date and time
Date() function, in its simplest form, will return current date and time as displayed on user computer.

var localtime= new Date();
// returns eg. Mon Nov 9 18:37:46 UTC+1100 2009

I often use a “combination” shown below to time stamp dynamic data (eg. frequently updated RSS feeds) in order to prevent caching so the information cannot be recalled from a local browser cache but is always called directly from the source:

var o= new Date();
var p= o.getTime();

Then I add dynamically extra parameter to source data URL :

var url= “somepage.html?p=”+p;

It is simpler than trying to get all “no-cache” headers and metatags to work consistently in all browsers.

Date and time in the past
By adding an additional parameter, the same Date() function can be used to convert a string of letters to a proper date format (ie. any date, expressed in “local time”):

var localtime= new Date(“Fri, 06 Nov 2009 04:56:00”);
// returns Fri Nov 6 04:56:00 UTC+1100 2009

Past date and time in GMT
Any past date expressed in local time can be easily converted to GMT.

var localtime= new Date(“Fri, 06 Nov 2009 04:56:00”);

var gmt=localtime.toUTCString();
// returns: Thu, 5 Nov 2009 17:56:00 UTC

GMT to local time
Reverse conversion from GMT to local time is also very easy. If your input time is expressed in GMT and you need local equivalent, just include “GMT” letters in the code to flag the input is not a local time!

var localtime= new Date(“Fri, 06 Nov 2009 04:56:00 GMT”); or
var localtime= new Date(“2009/11/05 04:56:00 GMT”);
// returns Fri Nov 6 15:56:00 UTC+1100 2009

Time in milliseconds (epoch)
Epoch is a very handy format for storing time info in a database and for all sorts of time related calculations.

var fulldate= new Date(); //returns local time as set on the user computer
var epoch=fulldate.getTime(); // converted to milliseconds

Unfortunately, I could not find any information on acceptable date formats. For example, this very frequently used Unix format does not work in JavaScript Date() function: “2009/11/06T03:42:40Z” - it has to be converted to a format that JavaScript can recognise before use.

Thursday, November 12, 2009

MashupAustralia highlights Pt 2

These are the last days of the competition so the number of submission is growing rapidly. Today just a few more examples of the most interesting entries. Again, applications using maps to present the information are dominating so, I will focus specifically on those.

The first application I would like to mention is another creation in Silverlight and Bing: GeoDemo (from the author of previously featured CrimeFinder). I must admit, I am very impressing with vector handling capabilities of Bing Map, even in FireFox browser! And I like the concept of changing boundaries to more detailed as the user zooms in closer and closer. Installing Silverlight is really no trouble but unfortunately not an easy task for institutional users (especially those in the public service) so, that may explain low popularity of this application. But there is a JavaScript version available as well and it also performs surprisingly fast (shown below)

There are also two new entries dealing with bushfires. Victoria: Fire Ready depicts locations of Fire Brigades and Police Stations in the vicinity of a chosen address. Distance circles of 10, 25 and 45km from the address are marked to allow users discern how quickly aid can arrive in case of emergency. The application also plots incidents reported by Victorian Country Fire Authority as well as current weather and wind conditions from the Bureau of Meteorology. Source information is very similar to what I use in my Bushfire Incidents map but the concept of the application and presentation is totally different. And much more popular judging by the number of votes!

Firemash is an attempt to match official announcements from NSW Rural Fire Services with incidents reported by a community via Twitter. Users can plot location of their houses on the map and if any incidents are reported in the vicinity, the application will retweet the information to alert the user. Interesting concept although its usefulness will be difficult to demonstrate at the moment because there are no fires reported in the entire NSW!

I will continue with more tomorrow…

A milestone to celebrate

This week I reached a small milestone – 1 million page impression served by (as measured by Google Adsense). It took me over 3.5 years to get to this number so, not a very spectacular result but… any excuse to celebrate is good enough for me! Especially considering I only have a handful of pages and not all carry Adsense ads. Not to mention that I practically do not advertise the site at all – so it is mostly just from organic traffic.

You probably wonder what this translates to in terms of revenue… Let me just say, it is enough for a long haul, first class overseas trip I look forward to so much! Again, considering that I get even as low as $0.01 per click on some ads, it is surprising how those cents quickly accumulate… And it appears not hugely out of line, comparing to what other publisher are able to get for their stock with comparable volume of visitors.

Speaking about Adsense advertising program. My biggest problem is to get relevant ads on my site. Serving weight loss ads on pages dealing with emergencies around the world or bushfires is not going to get me a good return! Generally, I have only limited text content so, Adsense automated choice of ads is less than optimal. I understand that Google tries to match keywords from the page with highest paying ad options. Seems logical and it probably works fine for pages with lots of text but this approach is a major limitation for sites like mine. For example, I had to remove any references to “fire” and “bushfire” from the front page because I was only getting ads on fire training and fire extinguishers… Probably best payout options for the page keywords but click through rate was appalling! By the way, it still does not explain why those weight loss ads are showing up on pages dealing with bushfires… Using tag does not help much either.

My site is about maps so Adsense ads are mainly about maps but again, this is less than ideal. Travel related information, in the broadest term, would be more relevant for my visitors but there is no way, at least not that I know of, to “force” specific ad content with Adsense program. I can understand why many publishers get frustrated with Google and try alternatives. My next big challenge will be to find those alternatives or maybe even to start my own advertising program. I dipped my toe in the “affiliate pool of money” earlier this year but so far the results are even worse than with Adsense…

Tuesday, November 10, 2009

Malocclusion: Disease of Civilization, Part V

Prenatal Development of the Face and Jaws

The structures of the face and jaws take shape during the first trimester of pregnancy. The 5th to 11th weeks of pregnancy are particularly crucial for occlusion, because this is when the jaws, nasal septum and other cranial structures form. The nasal septum is the piece of cartilage that forms the structure of the nose and separates the two air passages as they enter the nostrils.

Maternal Nutritional Status Affects Fetal Development

Abnormal nutrient status can lead to several types of birth defects. Vitamin A is an essential signaling molecule during development. Both deficiency and excess can cause birth defects, with the effects predominantly targeting the cranium and nervous system, respectively. Folic acid deficiency causes birth defects of the brain and spine. Other nutrients such as vitamin B12 may influence the risk of birth defects as well*.

The Role of Vitamin K

As early as the 1970s, physicians began noting characteristic developmental abnormalities in infants whose mothers took the blood-thinning drug warfarin (coumadin) during the first trimester of pregnancy. These infants showed an underdevelopment of the nasal septum, the maxilla (upper jaw), small or absent sinuses, and a characteristic "dished" face. This eventually resulted in narrow dental arches, severe malocclusion and tooth crowding**. The whole spectrum was called Binder's syndrome, or warfarin embryopathy.

Warfarin works by inhibiting vitamin K recycling, thus depleting a nutrient necessary for normal blood clotting.
It's now clear that Binder's syndrome can result from anything that interferes with vitamin K status during the first trimester of pregnancy. This includes warfarin, certain anti-epilepsy drugs, certain antibiotics, genetic mutations that interfere with vitamin K status, and celiac disease (intestinal damage due to gluten).

Why is vitamin K important for the development of the jaws and face of the fetus? Vitamin K is required to activate a protein called matrix gla protein (MGP), which prevents unwanted calcification of the nasal septum in the developing fetus (among
other things). If this protein isn't activated by vitamin K during the critical developmental window, calcium deposits form in the nasal septum, stunting its growth and also stunting the growth of the maxilla and sinuses. Low activity of MGP appears to be largely responsible for Binder's syndrome, since the syndrome can be caused by genetic mutations in MGP in humans. Small or absent sinuses are common in the general population.

One of the interesting things about MGP is its apparent preference for vitamin K2 over vitamin K1.
Vitamin K1 is found predominantly in green vegetables, and is sufficient to activate blood clotting factors and probably some other vitamin K-dependent proteins. "Vitamin K2" refers to a collection of molecules known as menaquinones. These are denoted as "MK", followed by a number indicating the length of the side chain attached to the quinone ring.

Biologically important menaquinones are MK-4 through MK-12 or so. MK-4 is the form that animals synthesize from vitamin K1 for their own use. Certain organs (brain, pancreas, salivary gland, arteries) preferentially accumulate K2 MK-4, and certain cellular processes are also selective for K2 MK-4 (
MGP activation, PKA-dependent transcriptional effects). Vitamin K2 MK-4 is found almost exclusively in animal foods, particularly pastured butter, organs and eggs. It is always found in foods designed to nourish growing animals, such as eggs and milk.

Humans have the ability to convert K1 to K2 when K1 is ingested in artificially large amounts. However, due to the limited absorption of normal dietary sources of K1 and the unknown conversion efficiency, it's unclear how much green vegetables contribute to K2 status. Serum vitamin K1 reaches a plateau at about 200 micrograms per day of dietary K1 intake, the equivalent of 1/4 cup of cooked spinach (see figure 1 of this paper). Still, I think eating green vegetables regularly is a good idea, and may contribute to K2 status.
Other menaquinones such as MK-7 (found in natto) may contribute to K2 status as well, but this question has not been resolved.

Severe vitamin K deficiency clearly impacts occlusion. Could more subtle deficiency lead to a less pronounced form of the same developmental syndrome? Here are a few facts about vitamin K relevant to this question:
  • In industrial societies, newborns are typically vitamin K deficient. This is reflected by the fact that in the US, nearly all newborns are given vitamin K1 at birth to prevent potentially fatal hemorrhage. In Japan, infants are given vitamin K2 MK-4, which is equally effective at preventing hemmorhage.
  • Fetuses generally have low vitamin K status, as measured by the activity of their clotting factors.
  • The human placenta transports vitamin K across the placental barrier and accumulates it. This transport mechanism is highly selective for vitamin K2 MK-4 over K1.
  • The concentration of K1 in maternal blood is much higher than its concentration in umbilical cord blood, whereas the concentration of K2 in maternal blood is similar to the concentration in cord blood. Vitamin K2 MK-7 is undetectable in cord blood, even when supplemented, suggesting that MK-7 is not an adequate substitute for MK-4 during pregnancy.
  • In rat experiments, arterial calcification due to warfarin was inhibited by vitamin K2 MK-4, but not vitamin K1. This is probably due to K2's ability to activate MGP, the same protein required for the normal development of the human face and jaws.
  • The human mammary gland appears to be the most capable organ at converting vitamin K1 to K2 MK-4.
Together, this suggests that in industrial societies, fetuses and infants are vitamin K deficient, to the point of being susceptible to fatal hemorrhage. It also suggests that vitamin K2 MK-4 plays a critical role in fetal and early postnatal development. Could subclinical vitamin K2 deficiency be contributing to the high prevalence of malocclusion in modern societies?

An Ounce of Prevention

Vitamin A, folic acid, vitamin D and vitamin K2 are all nutrients with a long turnover time. Body stores of these nutrients depend on long-term intake. Thus, the nutritional status of the fetus during the first trimester reflects what the mother has been eating for several months
before conception.

Dr. Weston Price noted that a number of the traditional societies he visited prepared women of childbearing age for healthy pregnancies by giving them special foods rich in fat-soluble vitamins. This allowed them to gestate and rear healthy, well-formed children.
Nutrient-dense animal foods and green vegetables are a good idea before, during and after pregnancy.

* Liver is the richest source of vitamin A, folic acid and B12.

** Affected individuals may show class I, II, or III malocclusion.

Monday, November 9, 2009

New entries in mashup contest

There are a few new entires in the MushupAustralia contest. With only a handful of exceptions they are all built with at least a simple mapping capabilities (you guessed it, overwhelmingly Google Maps!). Here is a brief description of the most interesting ones, in no particular order:

Fridgemate: simple concept of creating a "fridge list" and a map of places of interest in a local area. Currently the most popular mashup, with top score and the largest number of votes.

CrimeFinder: very impressive thematic mapping application built with Silverlight and Bing Maps. Nice user interface. What is of interest to me the most is how did the author managed to incorporate Local Government Boundaries in vector format. I know, I know, Silverlight is vector format (as Flash) but it appears there is some support as well for vector generalisation since boundaries are redrawn with different level of detail depending on the map zoom extents.

Suburban Trends: also very impressive thematic mapping application with ABS population statistics and various crime related information. Great use of dynamically loaded vector polygons of suburbs and Google Chart API.

NSW Crime Explorer: another application exploring crime statistics in tables, graphs and thematic maps (it appears to use static KMLs though).

ABS In Motion: statistics presented on an interactive Flash charts (not for low end computers).

Demographic Drapes: OpenLayers application using Google Maps as a backdrop and a number of thematic layers with populaution statistics and various administrative boundaries. More like a traditional online GIS.

Tonight I have also submitted my last application for this contest: revised Bushfire Incidents map. Just in time for the new bushfire season. I added extra controls to show additional information on the map:
- wind conditions (live from Bureau of Meteorology Web Feature Service)
- RSS feeds from Victorian CFA and NSW RFS (geocoded on the fly where possible)
- Locations of Fire Brigades in Victoria (over 1,200 points), and
- Weather widget and YouTube video player.

This is the last week of the competition. Entries close at 4pm, Friday 13 November. I am still hoping to see submissions from the commercial end of town but maybe they are all too busy chasing real projects…