We’ve switched things up for this episode, with Dr. Bryan Walsh asking the questions and me on the other side of the microphone.  We’re talking about our new Blood Chemistry Calculator – the product of lab data from tens of thousands of people and a machine learning algorithm called XGBoost.  The calculator analyzes a simple, inexpensive set of blood markers for patterns and immediately forecasts the probability that you’ve got any of a long list of deficiencies, overloads, and even infections - without directly testing for any of them.

Bryan and I discuss all the details, including the science behind the calculator, how you can use this tool to track progress over time, and how the calculator is a game-changer for practitioners.  If you’re ready to dive in and see what it can do for you, check out the calculator now.

[00:02:52] Chris's blood chemistry journey.

[00:04:11] Podcast with Dr. Bryan Walsh: Risk Assessment in the Genomic Era: Are We Missing the Low-Hanging Fruit?

[00:04:36] How Tommy looks at blood chemistry.

[00:06:18] Study: Hu, Frank B., Ambika Satija, and JoAnn E. Manson. “Curbing the diabetes pandemic: the need for global policy solutions.” Jama 313.23 (2015): 2319-2320.

[00:07:32] Decision tables, Functional Blood Chemistry seminar, Denver, March 2017.

[00:10:27] Machine Learning.

[00:11:10] Dogs vs Cats, Deep Convolutional Neural Network.

[00:15:05] Pima Indians dataset. Note there are just 768 instances in this dataset and not thousands (as I said in the audio). This is important because that’s still enough to build a reasonably accurate model using XGBoost.

[00:18:02] Elite Performance Program.

[00:18:55] GlycoMark.

[00:19:04] Podcast: Why You Should Skip Oxaloacetate Supplementation, Fueling for Your Activity and More!

[00:19:25] Ceruloplasmin, adiponectin.

[00:21:10] Required markers.

[00:21:56] Podcast: Health Outcome-Based Optimal Reference Ranges for Cholesterol, with Tommy Wood, M.D.

[00:22:05] RDW Study: Horne BD, May HT, Muhlestein JB, Ronnow BS, Lappé DL, Renlund DG, et al. Exceptional mortality prediction by risk scores from common laboratory tests. Am J Med. 2009;122: 550–558. Additional references: 1, 2.

[00:22:44] Out of pocket costs.

[00:23:07] The Blood Chemistry Calculator.

[00:23:25] Calculator forecast specifications.

[00:26:48] Binary classification vs logistic regression.

[00:28:44] Clinical decision-making in difficult patients.

[00:30:18] The clinical crystal ball.

[00:30:42] Who's it for?

[00:31:58] Fitness professionals.

[00:32:21] Monthly membership.

[00:35:12] The licensed clinician.

[00:36:34] Quicksilver tri-test.

[00:39:51] 7-minute analysis.

[00:41:10] Evidence-based reference ranges.

[00:41:34] bloodcalculator.com.

[00:42:41] Podcast: National Cyclocross Champion Jeremy Powers on Racing, Training, and the Ketogenic Diet.

[00:43:45] It's a good time to be a software engineer.

[00:44:15] XGBoost Study: Chen, Tianqi, and Carlos Guestrin. “Xgboost: A scalable tree boosting system.” Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. ACM, 2016.

[00:44:39] Fatty Liver Index. Study: Bedogni, Giorgio, et al. "The Fatty Liver Index: a simple and accurate predictor of hepatic steatosis in the general population." BMC gastroenterology 6.1 (2006): 33.

[00:45:23] Atherogenic Index of Plasma (AIP).

[00:45:42] Study: Horne BD, May HT, Muhlestein JB, Ronnow BS, Lappé DL, Renlund DG, et al. Exceptional mortality prediction by risk scores from common laboratory tests. Am J Med. 2009;122: 550–558.

[00:49:30] Sensitivity and specificity.

[00:50:31] Sparse data handling.

[00:52:52] Growth mindset.

[00:55:16] Specializing in Not Specializing TED Talk.