Free Clinical AI Briefs: https://www.healthit.com/signup?utm_source=podcast&utm_medium=audio&utm_campaign=pod_ep11 · Course: https://oncologyaiacademy.com/course?utm_source=podcast&utm_medium=audio&utm_campaign=pod_ep11

Two people can share a birthday and age very differently. Scientists call that difference biological age, and for the first time we can measure it, see what drives it inside our cells, and begin to change it.

In Episode 11, oncologist Jivesh Sharma, M.D., and Talia, our AI cohost, take a tour of the science: a map of what ages our cells (mTOR, AMPK, telomeres, senescent cells, DNA tags, mitochondria and inflammation), the clocks that now read organ age from a blood sample, the large GLP-1 trials showing protection for the heart, kidneys and liver, five strategies to watch (GLP-1 medicines, metformin, rapamycin, senolytics and cell reprogramming), and how AI is shortening every cycle of discovery.

Research cutoff: November 10, 2026. The predictions near the end are the host's own views, not established facts. The medicines discussed are prescription medicines with real side effects; decisions about them belong with you and your doctor.

Thank you for listening. Follow HealthIT.com for the next conversation. What should we explore next? Suggestions: https://www.healthit.com/feedback

Disclosure: This episode uses an authorized AI-generated version of Jivesh Sharma's voice and an AI cohost. Diagrams, scenarios and animations are labeled illustrations or study summaries, not patient data. Company-reported results are labeled as such. General education, not personal medical advice.

CHAPTERS
0:00 Same birthday, different bodies
0:46 Two kinds of age
1:31 A map of aging: what goes wrong inside our cells
3:22 Reading the clock: measuring biological age
4:30 Why measuring changes everything
4:59 The GLP-1 story: heart, kidneys, liver, brain
6:49 Five strategies to watch
8:53 How AI speeds it up
10:09 A checkup ten years from now
11:01 What to do today
11:41 Predictions
12:26 Thank you

KEY SOURCES (full list on the YouTube episode page)
López-Otín et al., Cell 2013 and 2023: hallmarks of aging
https://doi.org/10.1016/j.cell.2022.11.001
Oh et al., Nature 2023 and Nature Medicine 2025: organ ages from blood proteins
https://doi.org/10.1038/s41586-023-06802-1
Lincoff et al., NEJM 2023: semaglutide and cardiovascular outcomes (SELECT)
https://doi.org/10.1056/NEJMoa2307563
Perkovic et al., NEJM 2024: kidney outcomes (FLOW)
https://doi.org/10.1056/NEJMoa2403347
Sanyal et al., NEJM 2025: liver (ESSENCE)
https://doi.org/10.1056/NEJMoa2413258
Adamou et al., International Journal of Stroke 2024: stroke, eleven trials
https://doi.org/10.1177/17474930241253988
Cummings et al., Lancet 2026: semaglutide in early Alzheimer's disease
https://doi.org/10.1016/S0140-6736(26)00459-9
Bhatt, Pandey and Misra, Mechanisms of Ageing and Development 2026: metformin and telomeres
https://doi.org/10.1016/j.mad.2026.112241
Avsec et al., Nature 2026: AlphaGenome
https://doi.org/10.1038/s41586-025-10014-0
Loyal: LOY-002 FDA safety section accepted (company-reported)
https://loyal.com/posts/loy-002-tas-announcement

Free Clinical AI Briefs: https://www.healthit.com/signup?utm_source=podcast&utm_medium=audio&utm_campaign=pod_ep10 · Course: https://oncologyaiacademy.com/course?utm_source=podcast&utm_medium=audio&utm_campaign=pod_ep10

What if, for every year you lived, medicine gave you back more than a year of healthy life?

In Episode 10, oncologist Jivesh Sharma, M.D., and Talia, our AI cohost, explore longevity escape velocity and the technologies that could get us closer. Five capabilities are maturing at the same time, and AI is starting to link them into one loop: reading the body (genomes, single cells, proteins, wearables), understanding it (AlphaFold, genome models, virtual cells), writing corrections (protein design, CRISPR, base and prime editing, epigenetic editing, partial reprogramming), delivering them (lipid nanoparticles, personalized mRNA cancer vaccines) and measuring whether they worked (aging clocks).

The goal is healthspan: more good years, not just a longer life. Research cutoff: November 3, 2026. The predictions near the end are the host's own views, not established facts.

Thank you for listening. Follow HealthIT.com for the next conversation. What should we explore next? Suggestions: https://www.healthit.com/feedback

Disclosure: This episode uses an authorized AI-generated version of Jivesh Sharma's voice and an AI cohost. Molecular structures are real Protein Data Bank entries (1UBQ, 4OO8, 1IGT, 1KX5, 1GT0); diagrams, scenarios and animations are labeled illustrations, not patient data or study results. Company-reported results are labeled as such. General education, not personal medical advice.

CHAPTERS
0:00 Outrunning the clock
1:08 Healthspan, not just lifespan
2:08 Aging is the root of disease
3:21 The convergence loop
4:17 Read: seeing the body in detail
5:21 Understand: AI learns the language of biology
7:23 Write: editing biology
9:43 Deliver: reaching the right cells
10:57 Measure: closing the loop
11:58 A possible future for one patient
13:04 For everyone
13:36 Predictions and escape velocity
14:29 Thank you

KEY SOURCES (full list on the YouTube episode page)
CDC NCHS: Mortality in the United States, 2024 (life expectancy)
https://www.cdc.gov/nchs/products/databriefs/db548.htm
Garmany & Terzic, JAMA Network Open 2024: healthspan-lifespan gaps
https://pmc.ncbi.nlm.nih.gov/articles/PMC11635540/
López-Otín et al., Cell 2023: Hallmarks of aging
https://doi.org/10.1016/j.cell.2022.11.001
NHGRI: DNA sequencing costs
https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Costs-Data
AlphaFold Protein Structure Database
https://alphafold.ebi.ac.uk/
Google DeepMind: AlphaGenome
https://deepmind.google/blog/alphagenome-ai-for-better-understanding-the-genome/
Musunuru et al., NEJM 2025: patient-specific base editing
https://www.nejm.org/doi/full/10.1056/NEJMoa2504747
Ocampo et al., Cell 2016: partial reprogramming
https://doi.org/10.1016/j.cell.2016.11.052
Merck and Moderna: Phase 3 INTerpath-001 results (company-reported)
https://www.merck.com/news/merck-and-moderna-will-present-results-from-landmark-phase-3-interpath-001-trial-for-intismeran-autogene-plus-keytruda-pembrolizumab-at-esmo-2026-presidential-symposium/
Oh et al., Nature 2023: organ aging from plasma proteins
https://doi.org/10.1038/s41586-023-06802-1
Loyal: LOY-002 FDA safety section accepted
https://loyal.com/posts/loy-002-tas-announcement

AI is moving from answering questions to taking physical actions. What does that mean for healthcare?

In Episode 9, Jivesh Sharma, M.D., and Talia, our AI cohost, explore recent developments in hospital delivery robots, supervised blood collection, autonomous ultrasound acquisition and robot-assisted surgery. We explain the feedback loop behind physical AI: perception, registration, controlled movement and repeated observation.

The practical question is whether a completed delivery, scan or blood draw helps the next step of care happen. A research demonstration, an FDA authorization and a company rollout establish different things; none alone proves better patient outcomes or lower costs.

Research cutoff: October 27, 2026. This episode covers developments from July through October 2026.

Thank you for watching or listening. Follow HealthIT.com for future conversations; on YouTube, like this episode and subscribe. Which healthcare task would you like us to explore next? Suggestions: https://www.healthit.com/feedback

Disclosure: This episode uses an authorized AI-generated version of Jivesh Sharma's voice and an AI cohost. Synthetic imagery and explanatory animations are illustrations, not footage of actual devices, patients or study results. General education, not personal medical advice.

CHAPTERS
0:00 What changed in healthcare robotics
0:47 From AI answers to physical actions
2:06 How a robot senses and moves
3:15 Robots behind the scenes
4:15 The blood-draw example
4:58 Acquiring an ultrasound scan
5:59 Surgery: control and autonomy
7:00 What patients may notice next
8:04 Thank you and closing

PRIMARY SOURCES
FDA: Aletta robotic blood-draw authorization
https://www.fda.gov/news-events/press-announcements/fda-authorizes-first-its-kind-robotic-blood-draw-device
Google DeepMind: Gemini Robotics 2
https://deepmind.google/blog/gemini-robotics-2-brings-whole-body-intelligence-to-robots/
Google: Gemini 4 Argon for Robotics
https://developers.googleblog.com/gemini-4-argon-for-robotics/
Diligent Robotics: Moxi 2.0 rollout
https://www.diligentrobots.com/blog/diligent-robotics-a-serve-robotics-company-begins-rolling-out-moxi-20
NEURA Robotics: healthcare announcement
https://neura-robotics.com/neura-robotics-brings-physical-ai-to-healthcare/
Sun et al.: autonomous thyroid ultrasound (Nature Partner Journals, Digital Medicine)
https://www.nature.com/articles/s41746-026-03260-7
FDA: OTTAVA classification order
https://www.accessdata.fda.gov/cdrh_docs/pdf25/DEN250068.pdf
J&J: OTTAVA engineering and authorization announcement
https://www.jnj.com/media-center/press-releases/johnson-johnson-receives-fda-market-authorization-in-the-u-s-for-its-ottava-robotic-surgical-system
DARPA: surgical robotics competition
https://www.darpa.mil/news/2026/darpa-competition-surgical
FDA: MELODY remote ultrasound system (2017 background)
https://www.accessdata.fda.gov/cdrh_docs/pdf16/K161354.pdf

Imaging AI is expanding from individual findings to broader interpretation, three-dimensional scans you can question, and restoration of degraded images. Yet radiology services still struggle to meet demand.

Jivesh Sharma, M.D., and Talia, an AI cohost, examine recent research including RADAR, NV-Reason-CT, nnFoundation and AMIFound. We then connect technical capability to a different question: can it help more people receive useful care?

The central lesson is that faster scan interpretation does not automatically mean care arrives sooner. The episode distinguishes model benchmarks, reader studies, measured screening workload and completed follow-up, with visual explanations of where the next bottleneck can appear.

Thank you for watching or listening. Follow HealthIT.com for future conversations; on YouTube, like this episode and subscribe. What imaging task or access problem would you like us to examine next? Send suggestions at https://www.healthit.com/feedback.

This episode uses an authorized AI-generated version of Jivesh Sharma's voice and an AI cohost. General education, not personal medical advice. Conceptual illustrations and hypothetical pathway animations are labelled; they are not patient scans or clinical model outputs.

Chapters
00:00 The prediction and the paradox
00:46 RADAR: one scan, many questions
01:42 NV-Reason-CT: questioning a 3D scan
02:39 nnFoundation and the pace of progress
03:29 AMIFound: a clearer image can still be wrong
04:19 Why radiologists remain in demand
05:19 MASAI and the bottleneck in care
06:54 EAGLE: detection needs completed follow-up
07:59 What to watch next
08:56 Thank you and invitation

Primary references
- RADAR, Science: https://www.science.org/doi/10.1126/science.aec6129
- NV-Reason-CT: https://huggingface.co/nvidia/NV-Reason-CT
- NV-Reason-CT preprint: https://arxiv.org/abs/2609.27511
- nnFoundation: https://arxiv.org/abs/2609.26924
- AMIFound: https://www.nature.com/articles/s42256-026-01310-z
- MASAI: https://pubmed.ncbi.nlm.nih.gov/39904652/
- MASAI interval cancers: https://pubmed.ncbi.nlm.nih.gov/41620232/
- EAGLE: https://www.nature.com/articles/s41591-026-04656-4
- RCR workforce census: https://www.rcr.ac.uk/media/n1fjvrv4/rcr-2025-clinical-radiology-workforce-census-report.pdf
- ACR workforce economics: https://www.acr.org/Clinical-Resources/Publications-and-Research/ACR-Bulletin/2026/workforce-economics
- Hinton original recording: https://www.youtube.com/watch?v=2HMPRXstSvQ

When a cancer treatment stops working, what should we investigate next? Could the computing power behind recent AI research help answer that question?

Jivesh Sharma, M.D., and Talia compare AI-assisted mathematics, AlphaFold, AlphaGenome and Anthropic’s early ART enzyme-system findings. They ask what each advance establishes, where experiments are still needed, and how we would measure value for patients.

The episode distinguishes proposed mathematical results from accepted proofs, DNA prediction from gene editing, and a research announcement from evidence of clinical benefit. It also examines treatment resistance, flow cytometry and the laboratory bottleneck. Better outcomes, preserved quality of life and more healthy years are the goals; these examples do not establish those benefits.

Research cutoff: October 13, 2026. The outlook for the next few years is a forecast, not a validated patient-care workflow.

Thank you for watching and listening. Follow HealthIT.com, like this episode and subscribe on YouTube. Which hard healthcare problem should come next? https://www.healthit.com/feedback

Jivesh’s narration uses his authorized AI voice. Talia is an AI cohost. This is educational discussion, not individual medical advice.

Chapters
00:00 The question
00:35 Navier–Stokes and mathematical scrutiny
01:40 Manuscripts, withdrawals and proof checks
02:30 Checking the intended claim
02:57 What more computing power buys
03:46 AlphaFold and the Nobel recognition
05:02 AlphaGenome: prediction, not editing
06:08 ART: discovery still needs experiments
07:11 Treatment resistance and flow cytometry
08:12 The laboratory bottleneck
09:12 Value that matters to patients
10:10 From research to patient benefit
11:21 Thank you, follow and share your question

Sources
OpenAI: Navier–Stokes announcement, September 8: https://openai.com/index/navier-stokes-solution/
Clay: September 11 statement: https://www.claymath.org/year_type/2026/
Clay: current problem status and official formulation: https://www.claymath.org/millennium/navier-stokes-equation/
Buckmaster: concurrent mathematical work: https://cims.nyu.edu/~tristanb/statement.pdf
OpenAI: mathematics catalogue: https://github.com/openai/math
Catalogue corrections, October 7: https://github.com/openai/math/blob/main/history.md
Proof/formalization critique, October 6 preprint: https://arxiv.org/abs/2610.08144
AlphaFold, Nature 2021: https://www.nature.com/articles/s41586-021-03819-2
2024 Nobel Prize in Chemistry: https://www.nobelprize.org/prizes/chemistry/2024/summary/
AlphaGenome, Nature 2026: https://www.nature.com/articles/s41586-025-10014-0
DeepMind: AlphaGenome scope and efficiency: https://deepmind.google/blog/alphagenome-ai-for-better-understanding-the-genome/
Anthropic: early ART findings, September 23: https://www.anthropic.com/news/claude-discovers-novel-enzyme-system
ARUP project announcement, October 6: https://www.utahbusiness.com/press-releases/2026/10/06/arup-laboratories-artificial-intelligence-project-selected-redtail-ai-factory-challenge/

A headline says AI discovered a new drug. What has to happen before that discovery helps a patient?

Jivesh Sharma, M.D., examines ProteinTalks, a September 2026 Nature study of breast cancer cells' responses to drugs, and the published 2025 Phase 2a rentosertib trial in idiopathic pulmonary fibrosis. Laboratory and retrospective results do not prove that model-guided treatment improves survival. Commercial ties are disclosed.

Release-day update, October 6, 2026: Insilico Medicine announced on September 10 that the first patient was dosed in the GENESIS-IPF-3 Phase 3 trial. Starting a trial does not establish benefit. Rentosertib remains investigational. The narration examines the 2025 Phase 2a results. https://insilico.com/news/isn1009261-insilico-medicine-doses-first-patient-genesis-ipf-3

Research cutoff: October 5, 2026; source status rechecked October 6.

Thank you for listening and watching. Follow HealthIT.com and suggest a topic: https://www.healthit.com/feedback

Talia is an AI cohost. Jivesh's narration uses an authorized AI-generated version of his voice. General education, not personal medical advice. HealthIT.com is part of Jivesh's healthcare technology work and the Nexgen Precision portfolio.

Primary sources:

FDA drug-development process: https://www.fda.gov/patients/learn-about-drug-and-device-approvals/drug-development-process

Sun et al. Nature, September 9, 2026: https://www.nature.com/articles/s41586-026-11001-9

Westlake University: https://en.westlake.edu.cn/news_events/westlakenews/research/202609/t20260918_69874.html

Xu et al. Nature Medicine, 2025: https://www.nature.com/articles/s41591-025-03743-2

FDA January 2025 draft guidance concerns AI supporting regulatory decisions, rather than drug discovery itself: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-use-artificial-intelligence-support-regulatory-decision-making-drug-and-biological

00:00 The question behind the headline

00:46 A candidate still has to earn its place

01:59 September: a model of drug response

03:21 Early trials are not the finish line

05:12 What would count as useful progress?

06:21 Thank you and next conversation

September brought new AI models, voice capabilities, and tools that can take actions. Which changes deserve our attention in healthcare?

Jivesh Sharma, M.D., reviews selected September announcements from OpenAI, Anthropic, and Google. He explains differences in capability, cost, availability, and access, then considers how healthcare teams can test whether a tool improves the whole job.

The goal is clearer information for patients, more time for care, and research that remains verifiable. Better patient outcomes, quality of life, and a longer healthspan are goals to evaluate, rather than benefits established by these product announcements.

The video includes explanatory diagrams. This is a selected roundup, not a catalogue of every release. Research was checked October 3, 2026, for announcements made during September.

Thank you for listening and watching. Follow HealthIT.com for future conversations, and tell us what you would like us to explore next: https://www.healthit.com/feedback

Jivesh's narration uses an authorized AI-generated version of his voice. This episode is for general education, not personal medical advice. HealthIT.com is part of Jivesh's healthcare technology work and the Nexgen Precision portfolio.

Primary sources:

OpenAI ChatGPT release notes: https://help.openai.com/en/articles/6825453-chatgpt-release-notes

OpenAI API changelog: https://developers.openai.com/api/docs/changelog

GPT-6.1 Sol announcement: https://openai.com/index/introducing-gpt-6-1-sol/

Fable 5.1 and Mythos 5.1: https://www.anthropic.com/claude-fable-and-mythos-5-1

Opus 5.5: https://www.anthropic.com/claude-opus-5-5

Sonnet 5.5: https://www.anthropic.com/claude-sonnet-5-5

Gemini 3.8 Flash and Flash Cyber: https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/

Gemini 4 Argon: https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-4-argon/

Gemini 3.8 Live: https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-gemini-3-8-live-extended-thinking/

Life Sciences Verification Program: https://www.anthropic.com/news/life-sciences-verification-program

Chapters:

00:00:00 Opening

00:00:48 OpenAI: longer work, lower-cost options

00:02:07 Anthropic: different models for different jobs

00:03:14 Google: a workhorse model and a restricted frontier release

00:04:11 Beyond models: voice and agents

00:05:17 Healthcare and research: access is not validation

00:06:32 What would make these updates useful?

00:07:20 What to watch in October and closing

An AI router can send a patient message to the right queue and still miss the symptom that matters.

Jivesh Sharma, M.D., and Talia, our AI cohost, examine message routing, emergency triage, prior authorization and breast screening. Animated diagrams explain how software allocates clinical attention, when it should escalate, and what to measure beyond accuracy.

Share a question or study at https://www.healthit.com/feedback. Please leave out patient-identifying information. Thank you for watching and listening.

CHAPTERS
00:00 Not urgent is not the same as not important
01:18 Deciding where things go
02:34 The least expensive intelligence that can safely do the job
04:14 Route by consequence
05:02 Triggers, and which way to err
05:46 Layers, not a single filter
06:11 Consumer triage
07:33 Hospital alerts and missed cases
08:11 The label was cost, not need
09:05 Prior authorization and incentives
10:40 Human review
11:28 How much attention, not whether
12:29 What made it work
13:39 Same risk, different design
14:06 Five rules for an AI router
15:13 Pay for accuracy
15:38 Put the most important thing first
16:21 Not urgent should never mean not important

SOURCES AND SCOPE
OPTIC (preprint; agreement with GPT-4 labels, not patient outcomes): https://arxiv.org/abs/2503.05701
Johns Hopkins team overview: https://bids.jhmi.edu/news/triaging-patient-messaging-with-ai/
ChatGPT Health triage test, Nature Medicine (2026): https://www.nature.com/articles/s41591-026-04297-7
Methodological counterpoint (preprint): https://arxiv.org/abs/2603.11413
Earlier Epic sepsis model, retrospective validation (2021): https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2781307
Cost-proxy bias (2019; simulated change): https://doi.org/10.1126/science.aax2342
Medicare WISeR: https://www.cms.gov/priorities/innovation/innovation-models/wiser
CMS timing and appeal-payment details: https://www.cms.gov/priorities/innovation/files/document/wiser-model-frequently-asked-questions
EFF release and corrections: https://www.eff.org/deeplinks/2026/09/new-records-reveal-problems-medicares-ai-prior-authorization-experiment
CMS records via EFF: https://www.eff.org/document/wiser-foia-response-combined-records
Jessica Fu, The Seattle Times, Sept. 16, 2026 (syndicated; vendor denominators and response): https://www.arcamax.com/business/businessnews/s-4299526
Early Washington and Arizona rates are descriptive, not adjusted comparisons of accuracy or clinical appropriateness. The 83-day delay covers the full request pathway; CMS's 72-hour target applies to direct vendor submissions. EFF is an advocacy organization that challenged the program.
MASAI interim analysis (2023): https://pubmed.ncbi.nlm.nih.gov/37541274/
MASAI interval-cancer analysis (2026): https://pubmed.ncbi.nlm.nih.gov/41620232/
MASAI studied one Swedish workflow against double reading. Non-inferiority does not establish that automation bias cannot cause errors.

DISCLOSURES
The patient example is fictional. Jivesh Sharma's voice is AI-generated with his authorization; Talia is an AI cohost. Jivesh Sharma, M.D., is CEO of Nexgen Precision, a healthcare AI venture studio. No Nexgen Precision product is discussed. Proposed routing rules are editorial recommendations, not a validated clinical protocol.
Research cutoff: September 28, 2026. Sources checked September 30, 2026.