Jump to content

User talk:Doc James

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Note: Mostly over at MDWiki.org

What do you think of the BRI vs. BMI? Also, this.

We don't have an article on BRI yet, but it's getting more public attention. Might be worthwhile to include in your calculators project. Zefr (talk) 01:28, 21 September 2024 (UTC)[reply]

User:Zefr Yah for sure it would be reasonable to have that calculator aswell. Do you want to give it a try building it? It is not that complicated. Doc James (talk · contribs · email) 02:07, 21 September 2024 (UTC)[reply]
Sure. Will work at it over the next few days, and will alert you here. Zefr (talk) 02:18, 21 September 2024 (UTC)[reply]
A start class article is completed. Would benefit from your review and edits. Zefr (talk) 02:29, 24 September 2024 (UTC)[reply]
Amazing. You want to build a calculator to go in the article? Doc James (talk · contribs · email) 02:09, 25 September 2024 (UTC)[reply]
Using this format, I drafted an equation and tried to solve my own BRI, but it looked and felt clumsy.
When you say "build", how does that work? It would be a significant public contribution for Wikipedia to provide a calculator. The one in the article is easy to use, but is owned by a commercial enterprise. The one offered in the Thomas article is defunct.
For the article, I copied and cited the equation from the 'BRI definition' in the Zhang article, but the original equation development is more extensive and calculus-based under 'Supplementary material' (downloadable file) in the Thomas article. Zefr (talk) 02:42, 25 September 2024 (UTC)[reply]
User:Zefr Okay here is a draft User:Doc James/BRI. Will need some work to get it to function with metric and imperial like we have for BMI. Also need to figure out how to get two significant figures after the decimal point. Doc James (talk · contribs · email) 03:24, 25 September 2024 (UTC)[reply]
The variables presented in the calculation in your draft and in the article (from the Zhang source) are waist circumference and height, but the original Thomas calculation (from the publication under Web-based Calculator states: Screen shot of Body Roundness Calculator. After the user enters personal information for age, height, weight, sex, race, waist and hip circumferences, the program outputs total percent body fat, amount and percent of VAT, the body roundness index, and whether the individual is within the determined healthy range of body roundness. A graph depicting the resulting individual ellipse (African American curve) and healthy range (green shaded area) provide the user a visual representation of their body roundness relative to the healthy range. (my underlining).
In the Discussion, Thomas summarizes: "we may have reached the limits of maximal accuracy of predictions of VAT and % body fat that rely solely on the circumference measurements (waist and hip), age, height, sex, race, and weight."
The easy-to-use calculator here uses these same 6 variables (race/ethnicity is optional).
I explained in the article under Calculation: "The only device needed to determine BRI is a measuring tape for circumference of waist and hips, and determination of height," which comes from the sources.
Thoughts on best dealing with these variations for the BRI calculator? Zefr (talk) 18:47, 25 September 2024 (UTC)[reply]
That calculator provide Percent Body Fat, BMI, and BRI. Depending which one of those you want you need to provide different variable. For BRI you only need height and waist. And that is all my BRI calculator uses. Doc James (talk · contribs · email) 18:56, 25 September 2024 (UTC)[reply]
Anyway the calculator can also be adjusted to take imperial as we see here mdwiki:Template:BMI calculator2 Doc James (talk · contribs · email) 00:25, 26 September 2024 (UTC)[reply]

User:Zefr Built another one here mdwiki:Pulmonary_embolism#Probability_testing Doc James (talk · contribs · email) 00:57, 26 September 2024 (UTC)[reply]

Both the BMI and PE calculators are nice. What is needed to finish the BRI calculator? Zefr (talk) 01:07, 26 September 2024 (UTC)[reply]
Would be nice if it accepted metric or imperial, similar to the BMI calculator. Doc James (talk · contribs · email) 01:18, 26 September 2024 (UTC)[reply]

Another one at mdwiki:Chest_pain#Risk_scores. Same data that EN WP provides but is interactive and calculates a result based on user imputs. Doc James (talk · contribs · email) 02:04, 26 September 2024 (UTC)[reply]

Another good one. Given your experience in building the BMI calculator, can't that one be the model to complete the BRI? I wasn't following the VP-Tech discussion you had earlier this month - are there editors from there that could finish the BRI?
As I didn't have VP-Tech on my watchlist, and didn't know about your discussion, I posted a question there today. Apologies for the replication. Zefr (talk) 02:42, 26 September 2024 (UTC)[reply]
User:Zefr tool takes a bit of trial and error. Have the inputs formating correctly for metric and imperial here User:Doc James/BRI2 but just struggling to get the formula to work... Doc James (talk · contribs · email) 03:22, 26 September 2024 (UTC)[reply]
User:Zefr okay have it working now. Will need to get the tool activated by default to have it be generally useful. That will require consensus at WT:MED and VPT. Doc James (talk · contribs · email) 15:16, 26 September 2024 (UTC)[reply]
Does look good, thanks. The equation in the article was rewritten with this interesting history.
Uwappa had some thoughts about color-coding the cells. What do you think?
As you are working on several calculators, are you going to provide a list for WT:MED to review? VPT seems it's had enough discussion about calculators for now. Zefr (talk) 15:45, 26 September 2024 (UTC)[reply]
We have both formating styles for BMI results mdwiki:Body mass index. IMO no need to limit oneself to one.
Yes working on a bunch of calculators for MDWiki. If folks here wish to use some of them they are more then welcome to. All material licensed under a compatible license. Doc James (talk · contribs · email) 15:51, 26 September 2024 (UTC)[reply]
What do you recommend next for the BRI calculator being in the article? Should it be applied as for BMI, or was there something further you'd prefer to tweak or add? Zefr (talk) 16:40, 26 September 2024 (UTC)[reply]
User:Zefr IMO I think it is ready to go in the article. But we would need to get it turned on for general use. Doc James (talk · contribs · email) 17:33, 26 September 2024 (UTC)[reply]
Done, with thanks for your quick work. As a fairly wide box, it seems practical in the white space beside the TOC, knowing this may violate a style rule. Thoughts? Zefr (talk) 18:01, 26 September 2024 (UTC)[reply]
Placed in the infobox... Let me know what you think?
Also it does not work unless people have turned the gadget on. So we either need to build a notice for those who do not have it on how to turn on the gadget, or get it activated by default.
Asked at User_talk:Xaosflux#Updating Doc James (talk · contribs · email) 18:11, 26 September 2024 (UTC)[reply]
Good point about the gadget - most visitors to the page would not know otherwise, and may be frustrated by an unsolved BRI (IP users especially).
Concerning the BRI for % total fat and % visceral adipose tissue, Thomas says "maximal accuracy of predictions of VAT and % body fat rely solely on the circumference measurements (waist and hip), age, height, sex, race, and weight", indicating another calculator including the additional 5 variables - hip, age, gender, ethnicity, weight - would offer something more (possibly only for researchers)... or does this get too deep into the weeds? Zefr (talk) 18:39, 26 September 2024 (UTC)[reply]
One could build other calculators for % total fat and % visceral adipose tissue, these are different than BRI though. Doc James (talk · contribs · email) 19:13, 26 September 2024 (UTC)[reply]
Good points. Seems best for the widest viewership to remove it until a fix or disclaimer is provided. Thoughts? Zefr (talk) 01:34, 27 September 2024 (UTC)[reply]
Good to have something for people to review and provide feedback on. I would leave it for now and we will continue to work on it. Doc James (talk · contribs · email) 02:44, 27 September 2024 (UTC)[reply]
Looks like issues are now addressed Doc James (talk · contribs · email) 03:49, 27 September 2024 (UTC)[reply]

Discussion now at Wikipedia:Village_pump_(technical)#Turning_on_the_Calculator_gadget_as_default Doc James (talk · contribs · email) 18:33, 26 September 2024 (UTC)[reply]

Thought 1: Provide realistic default values.
  1. The calculator will show a meaningful result from the start, even to readers who do not input numbers.
  2. Meaningful default input examples are more inviting to enter real data. Also it will require less clicks to step up/down, compared to clicking all the way from zero.
  3. Exceptions like "NaN" and "infinity" will be rare.
Thought 2: Compute a healthy target. Given the input height, what range should the weight or waist be to score "green"? E.g. for a height of 180cm, a 'green' weight should be in the range 60-81kg. Uwappa (talk) 15:45, 28 September 2024 (UTC)[reply]
Yah default values are already possible. What would you like them to be. Doc James (talk · contribs · email) 16:01, 28 September 2024 (UTC)[reply]
Any realistic numbers will do. So, just go for something close to the worldwide average? Numbers found using Wolphram Alpha:
  • height: 162 cm
  • weight: 69 kg
And update those defaults later when anyone comes up with a better idea? Uwappa (talk) 16:25, 28 September 2024 (UTC)[reply]
BRI uses waist and height... have added defaults[1] Doc James (talk · contribs · email) 02:37, 29 September 2024 (UTC)[reply]
The BRI article has had some changes and considerable talk page discussion. It would benefit from your review and editing, with thanks. Zefr (talk) 19:27, 12 October 2024 (UTC)[reply]
Hey Zefr. Mainly just interested in the calculator aspect of things. Best Doc James (talk · contribs · email) 03:02, 13 October 2024 (UTC)[reply]

Impact on calculator, show a silhouette matching the computed BRI, with a background colour for health level.

Step 1, The default situation

Calculator
Units 1 Metric
0 Imperial
Height 178 cm5 feet 10.078740157480311 in
Waist 80 cm31.496062992125985 in
BRI
body_roundness_index_silhouettes.svg
2.46

The silhouette is lean. The background colour is yellow, signalling a warning.

The silhouette should be in the center of the cell. Ideally the computed BRI should be on the chest. See Template_talk:Annotated_image#Minimum_width_for_crop? for a reported problem with cropping to narrow images.

Step 2, Reader enters own height, e.g. 160 cm. The silhouette gains some width, the computed BRI value changes, the background colour changes towards green.

Calculator
Units 1 Metric
0 Imperial
Height 160 cm5 feet 10.078740157480311 in
Waist 80 cm31.496062992125985 in
BRI
body_roundness_index_silhouettes.svg
3.36

Step 3, Reader enters own waist, e.g. 116 cm. The silhouette changes to an overweight silhouette. The background colour changes to a warning colour.

Calculator
Units 1 Metric
0 Imperial
Height 160 cm5 feet 10.078740157480311 in
Waist 116 cm31.496062992125985 in
BRI
body_roundness_index_silhouettes.svg

8.57

Step 4, Reader plays with waist size. What should my waist be to be green, healthy? A waist of 93cm shows a healthy silhouette in green.

Calculator
Units 1 Metric
0 Imperial
Height 160 cm5 feet 10.078740157480311 in
Waist 93 cm31.496062992125985 in
BRI
body_roundness_index_silhouettes.svg

5.01

Step 5, Reader takes action in real life to bring down waist size down from 116 to 93cm. Wikipedia has helped to improve the health of a reader. Good!

Colours, BRI value ranges and the silhouette are still work in progress, so please do not change the calculator yet. What you could do to help with your technical skills:

  1. Put choices Metric and Imperial below each other, left aligned. It will read faster, in one eye fixation. Also it will suit the silhouettes better.
  2. How to get the silhouette in the center, above the BRI number?
  3. Have a look at Template_talk:Annotated_image#Minimum_width_for_crop? I am not sure if a BRI number with 2 decimals will look nice on the chest of lean silhouettes, but it's still good to fix a bug.
  4. Check the proportions of the silhouettes. They look OK to me and I do trust user:Cmglee's computations, but just check to be sure.
  5. Define BRI colours in constants, e.g. BRI1 to BRI20, so there is one set of colours for , the calculator, the 'graph' in BRI article and possible other future uses in the future such as articles in other languages
  6. Just a nice to have: intermediate colours for the calculator, e.g. a computed colour code for BRI values 1.62, 1.71, 1.85 somewhere between the fixed colours of BRI1 and BRI2. The calculator could show subtle colour changes for every cm change of waist.

Uwappa (talk) 09:25, 13 October 2024 (UTC)[reply]

Wow, amazing... Doc James (talk · contribs · email) 16:13, 13 October 2024 (UTC)[reply]
Thank you. Please assist with the programming at: Template:Body_roundness_index/sandbox Uwappa (talk) 18:24, 16 October 2024 (UTC)[reply]
Currently travelling. What is the top ask that is left to be completed? Doc James (talk · contribs · email) 21:20, 16 October 2024 (UTC)[reply]
Just one question left:
How to show only one of 20 silhouettes, based on variable roundedBoundedBRI?
Current path, almost there: have 20 BRI row, show only one row, based on roundedBoundedBRI.
I did try an alternative which failed: dynamically 'compute' a CSS class for just one row with 20 cropped images, e.g. formula = 'bri' + roundedBoundedBRI resulting in CSS class values like 'bri1', 'bri2', ... 'bri20'. Such CSS classes could take care of the background colour and showing only one cropped image in just one row. That idea did not work as I was unable to concatenate strings and don't know how such a dynamic string value could end up in a class= parameter.
Enjoy your trip! Uwappa (talk) 06:24, 17 October 2024 (UTC)[reply]
Will see if Brian has solutions... Doc James (talk · contribs · email) 17:02, 17 October 2024 (UTC)[reply]
And... found a solution after working on the calculator documentation, hoping that it would yield a new idea. And yes, it did!
  1. Use 20 hidden checkboxes for roundedBoundedBri values 1 to 20.
  2. Dynamically check only one of them using ifequal in formulas. See it working in the sandbox.
  3. todo, last hurdle: show only one BRI row using CSS, similar to what you already did with unit radio buttons. I've asked user:cmglee to do that last bit as my CSS skills are a bit rusty and slow.
This is very good news. We now can have the same type of calculator for BMI and Waist-to-height_ratio.
Some questions that require medical knowledge:
Uwappa (talk) 16:58, 17 October 2024 (UTC)[reply]
BRI and Waist to height take the same values so yes similar. But I think deserve separate pages though there could be a short overview pages. Doc James (talk · contribs · email) 17:08, 17 October 2024 (UTC)[reply]

Merge

[edit]

Hello! Should Unstable DNA sequence be merged into Genome instability? Polygnotus (talk) 16:41, 13 October 2024 (UTC)[reply]

Not sure... Doc James (talk · contribs · email) 18:01, 13 October 2024 (UTC)[reply]
Thank you. Perhaps a redirect would be better? I dunno, its pretty far outside my wheelhouse. Also, have you seen Wikipedia:Teahouse#Medical_article_creation_for_students_-_Topic_selection_and_audience_level? They are setting these students up to fail because they get assigned topics that already have an article. 4 people including myself have tried to explain that writing MEDRS articles is not a good beginner task. Perhaps you can help? I still have nightmares from the time I cleaned up after a WikiEdu course. Polygnotus (talk) 18:14, 13 October 2024 (UTC)[reply]
Mostly over at mdwiki:Main Page in part as EN WP is not supportive / actively prevents the development of multilingual content.[2] We have managed to rebuild the workflow without EN WP though.[3] Agree that use of AI would be a concern. Doc James (talk · contribs · email) 19:08, 13 October 2024 (UTC)[reply]
You know a lot about medicine, I (unfortunately) have spent an unhealthy amount of time on a computer. I put my trust in a doctor, you put your trust in a nerd. I understand feeling discouraged, and you obviously trust the person who said the only option was to fork, but as a nerd I have to say that that decision didn't make much sense, although I do understand the emotion surrounding it. Sometimes people get stuck, having one solution to a problem in mind, and then when that one solution doesn't work out they get really discouraged and see no other option than to leave. But usually there are many solutions to a problem. In this case the BRfA was doomed to fail, and it was silly to even attempt it, but that didn't actually matter because there were other (better) ways to achieve your goals. A "my way or the highway"-mentality rarely works out well on Wikipedia. Since that was apparently not their only mistake it may be wise to reconsider how much trust you put in which person. Polygnotus (talk) 19:44, 13 October 2024 (UTC)[reply]
That was just one among many reasons. I have made nearly 50,000 edits to MDWiki since mostly moving their 4 years back. I find I am now much more productive.
Plus those who understand English have tons of excellent medical resources to choose from. Those who do not have few or none. As a global public health effort I prefer to concentrate on the later. By the way always happy to have more good faith collaborators their :-) Doc James (talk · contribs · email) 20:29, 13 October 2024 (UTC)[reply]
Well, I hope you forgive me for not doing a deep dive into all the reasons. I just think that particular one is not strong. I was impressed by the amount of work that has been done on mdwiki. I understand English (on a good day) but not medicine. Polygnotus (talk) 05:07, 14 October 2024 (UTC)[reply]
No worries. Yah classes of students are a LOT of work... I am around here a bit but try to stay out of controversies. Doc James (talk · contribs · email) 05:39, 14 October 2024 (UTC)[reply]