What can a 2017 colonoscopy study teach us about using AI diagnostics safely in 2025?

An AI diagnostic tool boasts 99% accuracy. Should you trust it? In this episode, I explain why that number can be dangerously misleading and equip medical professionals with the practical strategies needed to see through the hype and protect their patients.As artificial intelligence becomes more integrated into healthcare, the ability to critically evaluate these tools is no longer optional; it's a core clinical skill. This session moves beyond the headlines to uncover the common, often hidden, flaws in AI training that can lead to inflated performance metrics and real-world risk. Learn how to become the essential human-in-the-loop who can distinguish a robust, reliable AI from a brittle and dangerous one.In this video, you will learn:The "Memorizing Student" Problem: A simple analogy to understand Overfitting, one of the most common ways AI models fail in the real world.How to Spot the Flaws: Practical techniques to diagnose unreliable AI, including how to interpret learning curves and why true external validation is the gold standard.The Danger of "Cherry-Picking": How selective reporting creates a false perception of reliability and why demanding transparency is crucial.The Colonoscopy Analogy: A powerful, real-world framework for how clinicians should approach AI results right now. Learn how to use a "positive" AI signal to your advantage and, more importantly, how to handle a "negative" signal to prevent catastrophic errors from automation bias.Your Ultimate Responsibility: Why the physician, not the algorithm, is always accountable and how to use AI as a tool for support, not an absolution of your clinical judgment.If you are a physician, medical student, resident, or healthcare administrator, this presentation provides the foundational knowledge you need to navigate the next wave of medical technology safely and effectively.

Are we caught in another tech hype cycle with AI? The promise of technology often clashes with financial and practical reality. This presentation cuts through the noise to offer a clear and balanced framework for thinking about the future of artificial intelligence in business and society.Join me for a thoughtful exploration of how to navigate the booms and busts of innovation. Using the "Could, Should, Might, and Don't" mindsets developed by futures designer Nick Foster, this talk dissects the different ways we approach the future—from the exciting, sci-fi visions of "Could Futurism" to the critical, ethical questions of "Don't Futurism."In this video, you will learn:How to recognize the phases of a technology hype cycle and avoid common pitfalls.The four key mindsets for engaging with the future of AI.Why user-centric design and adaptability are crucial for successful innovation, learning from past tech implementations.The importance of balancing optimism with a healthy dose of skepticism and ethical consideration.What are your thoughts on the future of AI? Which mindset—Could, Should, Might, or Don't—resonates most with you?

This episode provides an overview of Total Energy Expenditure (TEE) and Basal Energy Expenditure (BEE), emphasizing their definitions, measurement methods, and patterns across the human lifespan. It explains that TEE is the total of all calories burned, comprised mainly of BEE (the energy needed for basic functions at rest), the Thermic Effect of Food, and Physical Activity Energy Expenditure. Contrary to common beliefs, I explain that metabolism remains stable from ages 20 to 60, and midlife weight gain is primarily linked to lifestyle shifts rather than a sudden slowdown. The episode stresses the importance of objective measurement techniques, such as doubly labeled water for TEE, and highlights that body composition, particularly lean mass, is the main determinant of BEE. Finally, it corrects misconceptions regarding the impact of exercise and notes that metabolism declines gradually only after age 60, largely due to loss of lean tissue.

Sources:

Herman Pontzer, Yosuke Yamada, Hiroyuki Sagayama, Philip N. Ainslie, Lene F. Andersen, Liam J. Anderson, Lenore Arab, Issaad Baddou, Kweku Bedu-Addo, Ellen E. Blaak, Stephane Blanc, Alberto G. Bonomi, Carlijn V.C. Bouten, Pascal Bovet, Maciej S. Buchowski, Nancy F. Butte, Stefan G. Camps, Graeme L. Close, Jamie A. Cooper, Richard Cooper, et al. Daily energy expenditure through the human life course. Science, 373(6556):808–812, 2021. doi:10.1126/science. abe5017.

Kay Nguo, Helen Truby, and Judi Porter. Total energy expenditure in healthy ambulatory older adults aged ≥80 years: A doubly labelled water study. Annals of Nutrition and Metabolism, 79(2):263–273, 2023. doi:10.1159/000528872.

John R. Speakman, Jasper M. A. de Jong, Srishti Sinha, Klaas R. Westerterp, Yosuke Yamada, et al. Total daily energy expenditure has declined over the last 3 decades due to declining basal expenditure not reduced activity expenditure. Nature Metabolism, 5(4):579–588, 2023. doi:10.1038/s42255-023-00782-2.

Marie-Pierre St-Onge and Dympna Gallagher. Body composition changes with aging: The cause or the result of alterations in metabolic rate and macronutrient oxidation? Nutrition, 26(2):152–155, February 2010. doi:10.1016/j.nut.2009.07.004.

This episode challenges the common belief that weight gain in middle age is caused by a slowing metabolism, asserting that adult metabolism generally remains stable until after age sixty. Instead, it explains that weight gain is primarily driven by lifestyle factors, such as reduced physical activity and increased caloric intake. It details how the body mounts a powerful biological defense against weight loss, interpreting caloric restriction as a threat and triggering adaptive thermogenesis, which suppresses metabolism and increases hunger. Consequently, repeated cycles of weight loss and regain, known as weight cycling, become progressively more difficult and may disrupt metabolic health. The episode concludes that the most effective strategy for lifelong weight management is preventing weight gain through consistent lifestyle adjustments rather than relying on restrictive dieting after weight has accumulated.

Sources:

Herman Pontzer, Yosuke Yamada, Hiroyuki Sagayama, Philip N. Ainslie, Lene F. Andersen, Liam J. Anderson, Lenore Arab, Issaad Baddou, Kweku Bedu-Addo, Ellen E. Blaak, Stephane Blanc, Alberto G. Bonomi, Carlijn V.C. Bouten, Pascal Bovet, Maciej S. Buchowski, Nancy F. Butte, Stefan G. Camps, Graeme L. Close, Jamie A. Cooper, Richard Cooper, et al. Daily energy expenditure through the human life course. Science, 373(6556):808–812, 2021. doi:10.1126/science. abe5017.

Allyson K. Palmer and Michael D. Jensen. Metabolic changes in aging humans: current evidence and therapeutic strategies. The Journal of Clinical Investigation, 132(16):e158451, 2022. doi:10.1172/JCI158451.

Xiaotao Shen, Chuchu Wang, Xin Zhou, Wenyu Zhou, Daniel Hornburg, Si Wu, and Michael P. Snyder. Nonlinear dynamics of multi-omics profiles during human aging. Nature Aging, 4:1619–1634, 2024. doi:10.1038/s43587-024-00692-2.

Paul S. MacLean, Audrey Bergouignan, Marc-Andre Cornier, and Matthew R. Jackman. Biology’s response to dieting: the impetus for weight regain. Amer- ican Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 301(3):R581–R600, September 2011. doi:10.1152/ajpregu.00755.2010.

This episode fundamentally challenges the popular belief that human metabolism slows down during midlife, asserting instead that total energy expenditure remains stable between the ages of 20 and 60, only declining after age 60. Using gold-standard measurements, the text explains that the weight gain many adults experience is not due to a slowing metabolism but rather to changes in body composition, specifically the loss of lean muscle mass and the accumulation of fat. Furthermore, this episode highlights that aging and metabolic decline occur nonlinearly, with major molecular and functional transitions occurring around ages 44 and 60, which correspond to accelerated risks for age-related diseases. Consequently, this episode suggests that effective weight management strategies must focus on preserving muscle mass, promoting physical activity, and timing interventions to align with these critical metabolic inflection points.

Based on:

Herman Pontzer, Yosuke Yamada, Hiroyuki Sagayama, Philip N. Ainslie, Lene F. Andersen, Liam J. Anderson, Lenore Arab, Issaad Baddou, Kweku Bedu-Addo, Ellen E. Blaak, Stephane Blanc, Alberto G. Bonomi, Carlijn V.C. Bouten, Pascal Bovet, Maciej S. Buchowski, Nancy F. Butte, Stefan G. Camps, Graeme L. Close, Jamie A. Cooper, Richard Cooper, et al. Daily energy expenditure through the human life course. Science, 373(6556):808–812, 2021. doi:10.1126/science. abe5017.

Allyson K. Palmer and Michael D. Jensen. Metabolic changes in aging humans: current evidence and therapeutic strategies. The Journal of Clinical Investigation, 132(16):e158451, 2022. doi:10.1172/JCI158451.

Xiaotao Shen, Chuchu Wang, Xin Zhou, Wenyu Zhou, Daniel Hornburg, Si Wu, and Michael P. Snyder. Nonlinear dynamics of multi-omics profiles during human aging. Nature Aging, 4:1619–1634, 2024. doi:10.1038/s43587-024-00692-2.

This episode provides an overview of the significant challenge of dietary underreporting bias in nutrition research, especially concerning individuals with obesity. It explains how some people with higher body mass index (BMI) systematically and substantially under-report their true food and calorie consumption, often by 700 to 850 kcal per day, which is far more than their lean counterparts. The episode highlights that doubly labeled water (DLW) is the objective "gold standard" method for measuring actual calorie expenditure, and studies using DLW consistently validate the extent of this underreporting. This systematic inaccuracy, often driven by factors like social desirability bias, renders self-reported data from national surveys like NHANES largely unreliable for accurately assessing true calorie intake or its link to obesity trends. Ultimately, the episode cautions researchers to use objective measures and not rely solely on self-reported data when studying obesity.

Source on which the episode is based:

This episode provides an extensive critique of the reliability of self-reported dietary surveillance data, arguing that simple correlations between dietary trends (like increased calories or changes in macronutrients) and the rise in U.S. obesity rates are misleading. The episode emphasizes that correlation does not equal causation and highlights the significant problem of systematic underreporting of calorie intake, particularly among individuals with higher body mass index. Furthermore, I assert that observed increases in reported calorie intake over time may actually reflect changes in survey methodology rather than real shifts in eating behavior. Finally, I note that despite increasing use of dietary supplements, most Americans still fail to meet recommended intakes for several key micronutrients, underscoring that overall dietary quality remains suboptimal.

The studies on which this episode was based:

This episode provides an extensive overview of the complex biological regulation of appetite and energy balance, moving beyond the simple "calories in, calories out" model. It establishes that fat tissue is an active endocrine organ that produces hormones crucial for signaling the brain about the body's energy status. The episode highlights the central role of the hormone leptin, explaining that it suppresses hunger when fat stores are adequate and drives eating when stores are low. I contrast leptin deficiency, which causes unrelenting hunger and metabolic disease that is treatable with hormone replacement, with the more common condition of leptin resistance seen in obesity, where the brain fails to respond to high leptin levels, thereby promoting continued hunger and weight gain. Ultimately, this episode argues that appetite and weight regulation are governed by hormonal feedback systems that often override conscious control, underscoring the biological challenge of managing weight.

Studies on which this episode was based:

Dr. Shilpa Balaji Asegaonkar. Insights into role of adipose tissue as endocrine organ. International Journal of Diabetes Research, 1(1):01–04, January 2019. doi:10.33545/26648822.2019.v1.i1a.1.

Alexandros Vegiopoulos, Maria Rohm, and Stephan Herzig. Adipose tissue: between the extremes. The EMBO Journal, 36(14):1999–2017, June 2017. doi:10.15252/embj.201696206.

F. Lonnqvist. The obese (OB) gene and its product leptin–a new route toward obesity treatment in man? QJM, 89(5):327–332, May 1996. doi: 10.1093/qjmed/89.5.327.

Julie A. Chowen and Jesu ́s Argente. Leptin and the brain. HMBCI, 7(2):351–360, August 2011. doi:10.1515/hmbci.2011.113.

Jeffrey S Flier and Eleftheria Maratos-Flier. Obesity and the hypothalamus: Novel peptides for new pathways. Cell, 92(4):437–440, February 1998. doi:10.1016/s0092- 8674(00)80937- x.

Milen Hristov. Leptin signaling in the hypothalamus: Cellular insights and therapeutic perspectives in obesity. Endocrines, 6(3):42, August 2025. doi:10.3390/endocrines6030042.

Jiarui Liu, Futing Lai, Yujia Hou, and Ruimao Zheng. Leptin signaling and leptin resistance. Medical Review, 2(4):363–384, August 2022. doi: 10.1515/mr-2022- 0017.

This episode provides an extensive overview of the strong relationship between insufficient sleep and the increased risk of weight gain and metabolic dysfunction. It emphasizes that epidemiological data consistently identify short sleep duration as an independent risk factor for obesity, particularly in younger populations. Mechanistically, sleep deprivation is shown to disrupt appetite-regulating hormones, specifically by decreasing the satiety hormone leptin and increasing the hunger hormone ghrelin, while also impairing glucose metabolism and promoting insulin resistance. Furthermore, experimental studies confirm that restricted sleep leads to increased caloric intake, predominantly from snacks consumed during extended evening hours, resulting in a positive energy balance that favors fat storage. The episode concludes that because the relationship between sleep and obesity is bidirectional, addressing sleep health represents a crucial and modifiable public health strategy for managing metabolic risk.

The sources:

Sanjay R Patel and Frank B Hu. Short sleep duration and weight gain: a systematic review. Obesity, 16(3):643–653, 2008. doi:10.1038/oby.2007.118.

Shahrad Taheri, Ling Lin, Diane Austin, Terry Young, and Emmanuel Mignot. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Medicine, 1(3):e62, 2004. doi:10.1371/journal.pmed.0010062.

Karine Spiegel, Esra Tasali, Plamen Penev, and Eve Van Cauter. Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Annals of Internal Medicine, 141(11):846–850, 2004. doi:10.7326/0003-4819-141- 11- 200412070-00008.

Esra Tasali, Rachel Leproult, and Karine Spiegel. Reduced sleep duration or quality: relationships with insulin resistance and type 2 diabetes. Progress in Cardiovascular Diseases, 51(5):381–391, 2009. doi:10.1016/j.pcad.2008.10.002.

Arlet V Nedeltcheva, Jennifer M Kilkus, Jacqueline Imperial, Kristen Kasza, Dale A Schoeller, and Plamen D Penev. Sleep curtailment is accompanied by increased intake of calories from snacks. American Journal of Clinical Nutrition, 89:126–133, 2009. doi:10.3945/ajcn.2008.26574.

This episode provides an extensive overview of the strong relationship between insufficient sleep and increased risk of weight gain and metabolic dysfunction. It explains that large-scale epidemiological studies consistently show that short sleep duration is an independent risk factor for obesity, especially among children. The episode details the hormonal mechanisms through which sleep restriction promotes overeating, specifically noting the decrease in the satiety hormone leptin and the increase in the hunger hormone ghrelin. Furthermore, sleep loss impairs glucose metabolism and insulin sensitivity, creating metabolic conditions that favor fat storage and increase the risk of diabetes. Finally, the episode emphasizes the bidirectional nature of this issue, where obesity can worsen sleep quality, creating a vicious cycle of metabolic impairment.