Preventing Dialysis Access Failure: How Auvi Labs is Building a Wearable Ultrasound Device | Rishab Veldur & Kevin Volkema

In this episode, I sit down with Rishab Veldur (CEO) and Kevin Volkema (COO) from Auvi Labs to discuss how they're tackling a critical problem in dialysis care. 40% of dialysis patients experience fistula or graft failure within their first year—leading to endless hospitalizations and life-threatening situations.

Auvi Labs is building "Beacon," a wearable ultrasound patch that patients can use for just 10 minutes a day to detect early signs of access failure. We dive deep into their journey from a university capstone project to launching clinical pilots, the challenges of building a healthcare startup, and their unique approach to networking and building clinical partnerships.

Whether you're a healthcare founder, engineer, or just interested in medtech innovation, this episode is packed with actionable insights.

Timestamps

• 00:00 – Introduction

• 01:00 – The Problem: Dialysis Access Failure

• 03:15 – From Acoustic Device to Wearable Ultrasound

• 10:06 – The 20-Patient Pilot Study

• 11:52 – How to Find Clinical Pilots

• 17:40 – Networking Playbook for Healthcare Founders

• 21:01 – The Power of a Founder Newsletter

• 23:45 – Business Model & Value-Based Care

• 29:19 – Biggest Risks & Clinical Integration Challenges

• 36:04 – Bottlenecks for Scaling

Key Takeaways

1 Start with the problem, not the solution – They pivoted from an acoustic device to ultrasound after learning what patients and physicians actually needed

2 Healthcare is relationship-based – Cold outreach has low success; invest in building genuine connections over time

3 Skip the big conferences early on – Niche events and reaching out to researchers on Google Scholar yields better results

4 Use your student email – Everyone wants to help students

5 Send a newsletter – Share highs AND lows to bring people along on your journey

6 Value-based care is the path – Working with kidney contracting entities can bypass traditional CPT code reimbursement

Guest Links

• 🌐 Website: auvilabs.com

• 💼 Rishab Veldur (CEO) LinkedIn: https://www.linkedin.com/in/rishab-veldur/

• 💼 Kevin Volkema (COO) LinkedIn: https://www.linkedin.com/in/kevinvolkema/

Host

• 💼 Roupen Odabashian LinkedIn: https://www.linkedin.com/in/roupen-odabashian-md-frcpc-fasco-183aaa142/

• 📧 Email: roupen@deltahealth.tech

#healthcarestartup #medtech #dialysis #wearables #digitalhealth #medicaldevice #startup #healthcare #founders

Join us for an incredible conversation with Dr. Amit Phull, Chief Physician Experience Officer at Doximity, as we explore how the largest professional medical network in the United States is revolutionizing healthcare with AI-powered tools built by physicians, for physicians.

In this episode, we dive deep into:

- How Doximity grew to serve over 80% of U.S. physicians

- The development of HIPAA-compliant AI tools including Doximity GPT

- The importance of clinician input in healthcare technology

- Real-world impact: How AI is saving physicians hours and improving patient care

- The acquisition of Pathway (Montreal-based company) and integration of clinical decision support

- Privacy-focused AI scribe technology generating millions of notes

- Lessons learned from product failures and successes

Dr. Phull shares his unique journey from computer engineering to emergency medicine, and how maintaining both clinical practice and tech expertise positions him to bridge the gap between technology and healthcare delivery.

Timestamps

[00:00:00] Introduction & Welcome

[00:01:24] Dr. Phull's Journey: Computer Engineer to Emergency Physician

[00:04:00] Role as Chief Physician Experience Officer

[00:06:41] Why Maintaining Clinical Practice Matters

[00:11:04] Introduction to Doximity Tools & Platform

[00:14:52] The Secret Sauce: Clinician-Driven Development

[00:18:44] Doximity GPT Evolution & HIPAA Compliance

[00:22:21] Acquiring Pathway AI (Montreal)

[00:28:41] AI Scribe: Privacy-First Documentation

[00:30:52] Lessons from Product Failures

[00:36:18] Success Story: 10% to 90% Prior Auth Approval Rate

About Our Guest

Dr. Amit Phull, MD

Chief Physician Experience Officer, Doximity

Board-Certified Emergency Medicine Physician

Adjunct Lecturer, Northwestern University Feinberg School of Medicine

Dr. Phull combines deep expertise in medicine, technology, and strategy. He completed his MD at University of Virginia (where he also earned a BS in Computer Science), finished emergency medicine residency at Northwestern, and has been with Doximity since 2014. He continues to practice emergency medicine while leading physician experience strategy for the nation's largest medical professional network.

LinkedIn: https://www.linkedin.com/in/amit-phull-09931667/

Doximity Website: https://www.doximity.com

Doximity LinkedIn: https://www.linkedin.com/company/doximity

My email: roupen@deltahealth.tech

YouTube: https://youtu.be/ElB-a5Mubm4

Why do 30% of patients take medications differently than what's in their medical records? Dr. Eve Cunningham, Chief Medical Officer at Cadence, reveals the shocking gaps in traditional healthcare—and how remote patient monitoring is revolutionizing chronic disease management for 70,000+ patients across 20 health systems.

In this episode, we dive deep into:

✅ The hidden cost of episodic vs. continuous care

✅ How a practicing OB-GYN broke into healthtech leadership

✅ Why 20% of medication changes are actually DOWN-titrations

✅ The future of AI-powered clinical decision support

✅ Real outcomes: 18% fewer hospitalizations, $183/month savings per patient

TimeStamps:

Dr. Cunningham spent 20 years leading physician groups at Kaiser Permanente, CommonSpirit, and Providence before joining Cadence—a remote patient monitoring company backed by $141M from General Catalyst and Thrive Capital. She shares candid insights on physician leadership, technology transformation, and why clinicians MUST develop technical competency.

🎯 Perfect for healthcare entrepreneurs, medtech founders, physicians exploring innovation, and anyone building the future of digital health.

KEY TOPICS COVERED:

Remote patient monitoring at scale (70,000+ patients)

Clinical AI and machine learning in chronic disease management

Breaking into healthtech from clinical practice

Value-based care and Medicare reimbursement (CPT codes 99453, 99454, 99457, 99458)

Medication reconciliation and polypharmacy management

Virtual care infrastructure: telehealth, virtual nursing, hospital-at-home

Technology adoption in large health systems

The emerging clinician-engineer hybrid role

Deprescribing and down-titration opportunities

Social determinants of health and caregiver engagement

PUBLISHED OUTCOMES:

📊 New England Journal of Medicine: Catalyst validates Cadence's model: https://catalyst.nejm.org/doi/abs/10.1056/CAT.24.0521

CONNECT WITH DR. EVE CUNNINGHAM:

LinkedIn: https://www.linkedin.com/in/evecunninghammd/

🔗 CONNECT WITH CADENCE:

Company LinkedIn: https://www.linkedin.com/company/cadencerpm

Website: https://www.cadence.care/

Published Research: https://www.cadence.care/outcomes-report-2024

​MalkeAsaad, plastic surgery resident and founder of Rank RX, shares how he built an AI platform using large language models to revolutionize residency application screening. From med school in war-torn Aleppo to Mayo Clinic and MD Anderson, Malke discusses the unfair filtering system that inspired Rank RX—where Nobel Prize laureates get rejected for missing a cutoff by one point—and how AI can make hiring more objective and efficient.

What you'll learn:

- Why the current residency application system is broken (Nobel Prize winner rejected for 1-point score gap)

- How Rank RX uses AI/LLMs to screen 1,000–2,000 applications (30–80 pages each) in minutes

- Building a tech team as a physician entrepreneur without coding background

- Customer acquisition strategies for healthcare startups (networking, ads, vendor screening)

- Market validation: assessing if your solution solves a real problem people will pay for

Timestamps

- 0:00 – Unfair residency filtering: Nobel Prize winner rejected for 1-point gap

- 1:14 – Malke Assad’s journey: From Aleppo to leading U.S. institutions

- 3:45 – Rank RX: How AI/LLMs bring objectivity to application screening

- 4:21 – How it works: Custom scoring and program-driven selection criteria

- 8:36 – Real-world usage: Positive feedback and automated recommendation letter analysis

- 10:32 – Building a tech team without a coding background

- 17:35 – Key advice for physician entrepreneurs: Turning ideas into scalable companies

- RankRX Website: https://www.rank-rx.com/

- Malke Assad LinkedIn: https://www.linkedin.com/in/malke-asaad-43b908177

- The Match Guy Website: https://thematchguy.thinkific.com

When Sahil and his brother started Otoset in their mid-20s, they had no idea their FDA-cleared ear cleaning device would generate over 1 billion social media views and force them to completely pivot their business model.

In this episode, Sahil shares the unexpected journey from building a B2B medical device company to creating "the front door to ear care", a direct-to-consumer healthcare network serving 40 million Americans with chronic ear wax issues.

🔑 KEY TAKEAWAYS:

→ How they became some of the youngest founders to get FDA 510(k) clearance

→ The unexpected social media virality that changed everything

→ Why they pivoted from partner clinics to owning their own locations

→ Marketing strategies: organic content, influencers, and patient education

→ The critical role of FDA clearance as a competitive differentiator

→ Building credibility before scaling consumer marketing

→ Finding the right mentors in healthcare entrepreneurship

💡 WHO THIS IS FOR:

✓ HealthTech & MedTech founders navigating FDA pathways

✓ Startups exploring direct-to-consumer healthcare models

✓ Entrepreneurs learning to leverage social media for medical products

✓ Anyone interested in the consumerization of healthcare

📊 BY THE NUMBERS:

- 40 million Americans affected by ear wax buildup

- 1 billion+ views across social media

- $99 cash-pay model (first treatment)

- 20-30 patients/day in company-owned clinics

- Expanding to 50+ major metros

Timestamps:

00:00 - Introduction: The Brother's Ear Wax Problem

01:21 - What is Otoset? The First FDA-Cleared Ear Cleaning Device

03:22 - Why FDA Clearance Matters & How They Got It

06:56 - The Unexpected Social Media Explosion

08:55 - The Strategic Pivot: B2B Device to D2C Healthcare Network

11:36 - Business Model: $99 Cash-Pay & Building Owned Clinics

15:18 - Beyond Ear Care: Hearing Health & Expansion Plans

17:15 - Marketing Strategy: Organic, Influencers & Patient Education

20:33 - Building Credibility Before Scaling Consumer Marketing

22:35 - Biggest Lesson: Find Healthcare Entrepreneur Mentors Early

24:33 - Final Thoughts & Key Takeaways

📌 Key Resources & Links

🔗 Otoset Website: https://otoset.com/

🔗 Otoset Linkedin: https://www.linkedin.com/company/visitallears/

🏥 Find a Certified Clinic: https://otoset.com/pages/find-clinic

💼 Connect with Sahil: https://www.linkedin.com/in/sahildiwan/

Is AI really ready to replace doctors? Stanford PhD researcher Suana reveals shocking truths about medical AI that Big Tech doesn't want you to know. When she tested leading AI models like GPT-4, Claude, and DeepSeek on modified medical questions, their accuracy plummeted by up to 40%!In this eye-opening conversation, we dive deep into:

❌ Why 95%+ accuracy on medical exams means nothing in real clinical practice

❌ How AI models fail when there's "no right answer" (which happens constantly in medicine)

❌ The dangerous gap between flashy headlines and clinical reality

✅ How doctors can safely use AI as a co-pilot (not replacement)

✅ The future of medical AI evaluation and what needs to changeSuana is a 3rd-year PhD student at Stanford in Biomedical Data Science, pioneering real-world evaluation methods for medical AI. Her research on MedELM and benchmarking is reshaping how we think about AI deployment in healthcare.🔬

Key Research Discussed:

JAMA Open publication on AI robustness in medical diagnosis

MedELM: 35-dataset benchmark suite for real clinical tasks

Why MedQA and USMLE-style tests don't reflect actual patient care

⚠️ CRITICAL TAKEAWAY: AI models are trained to always give an answer, even when "none of the above" is correct—a potentially dangerous flaw in medical decision-making.📚 Resources Mentioned:

MedELM Leaderboard (public repository available)

Research on medical AI evaluation standards

Real-world hospital deployment considerations

Timestamps:

0:00 - Introduction: Why Medical AI Evaluation is Broken

1:04 - Suana's Journey: From Computer Science to Healthcare AI

2:32 - The 3 Critical Problems with Current AI Benchmarks

8:28 - The Research: Testing AI with "None of the Above"

17:24 - Shocking Results: AI Accuracy Drops 8-40%

19:02 - Why AI Can't Say "I Don't Know"

23:10 - Take-Home Message: Use AI as Co-Pilot, Not Replacement

24:58 - Real Clinical Examples: When AI Actually Helps

28:12 - MedELM: The Future of Medical AI Evaluation

34:35 - Final Advice for Doctors, Patients & Developers

Whether you're a physician, healthcare worker, AI developer, or patient curious about medical AI, this conversation will change how you think about artificial intelligence in healthcare.

Paper link: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2837372

Healthcare revenue integrity is broken, and Charlene, CEO of Ember, is fixing it with AI. In this episode, she reveals how hospitals lose millions due to inefficient billing processes and how her platform helps recover 51% of initially denied claims. From her background at Google Health to building a YC-backed startup that's generating over $60,000 per month in additional revenue for clients, Charlene shares the secrets behind focusing on one specific problem and executing flawlessly.

Key topics covered: Healthcare revenue cycle management, AI in medical billing, startup focus strategies, building in regulated industries, and the future of healthcare fraud detection. Perfect for healthcare executives, CFOs, startup founders, and anyone interested in healthtech innovation.

Timestamps

0:00 Introduction & Ember Overview

1:05 What is Revenue Integrity in Healthcare?

2:31 Ambient Listening & Clinical Documentation

4:10 Product Implementation & Integration Process

6:02 Real Results: $60K+ Monthly Revenue Recovery

8:33 Technical Deep Dive: How Ember Works

10:28 Charlene's Background & Healthcare Industry Insights

12:53 The $10B Healthcare Fraud Problem

16:17 Y Combinator Experience & Startup Focus

18:22 Building the Right Team (3-Year Journey)

21:44 Future Vision for Ember

22:41 Advice for Healthcare Entrepreneurs

Can AI forecast ICU risk from the first 36 hours of EHR data?

University of Washington researcher Sihan explains TrajSurv, a survival-prediction model that converts noisy, irregular ICU time series into interpretable latent trajectories using Neural Controlled Differential Equations (NCDEs) and time-aware contrastive learning aligned to SOFA. We cover how trajectories outperform snapshots, handle missingness without heavy imputation, and remain clinically legible via vector-field feature importance and trajectory clustering.

Validated on MIMIC-III and eICU with reported C-index ≈0.80 and cross-cohort ≈0.76, TrajSurv points to safer escalation, de-escalation, and bed allocation in the ICU.

In this episode: survival prediction basics; limits of Cox/RSF vs deep time-series models; NCDE explained in plain language; first-36h feature set (53 labs/vitals/demographics); metrics (C-index, Brier, dynamic AUC); interpretable clustering linked to outcomes; and what’s next—adding interventions for counterfactual simulation and extending to oncology.

Link to the paper: https://arxiv.org/abs/2508.00657

Timestamps

00:00 Why trajectories beat snapshots in EHR

01:00 Guest intro: Sihan, UW Biomedical Informatics

01:40 Survival prediction 101 and clinical use

03:40 From Cox/RSF to deep learning on time-varying data

05:03 What is TrajSurv (pronounced “traj-surf”)?

06:16 NCDE explained with the “ship + weather” analogy

08:14 Handling irregular sampling and missing data

09:14 Time-aware contrastive learning aligned to SOFA

10:47 Datasets: MIMIC-III and eICU; first 36h features (labs, vitals, demo)

12:40 Results: C-index ≈0.80; cross-cohort ≈0.76; interpretability

14:30 Workflow: CDS, monitoring, escalation, de-escalation

16:15 Why humans miss multi-variable long-horizon trends

18:21 Latent trajectory clustering and survival differences

23:18 Next: interventions, counterfactuals, oncology applications

25:40 Closing

Roupen Odabashian LinkedIn: https://www.linkedin.com/in/roupen-odabashian-md-frcpc-abim-183aaa142/

Sihang Zeng: https://www.linkedin.com/in/zengsh/

#HealthcareAI #ClinicalDecisionSupport #EHR #ICU #SurvivalAnalysis #DeepLearning #NCDE #MIMICIII #eICU #SOFA