🩺 AI Stethoscope | 15-Sec GP Checks # 64

15-sec heart checks, 20× leaner imaging, and smarter triage.

September 01, 2025

In partnership with

AIHealthTech Insider: Issue #64

September 01, 2025

AI is racing forward in healthcare again this week—bringing smarter diagnostics, more accurate imaging, and ethical pressure for clinical leadership.

Summaries for education, not medical advice. Verify locally before clinical use.

Image generated by AI

A UK trial shows an AI-enabled stethoscope from Imperial College London and Eko Health can rapidly detect heart failure, atrial fibrillation, and valve disease:

  • Analyzes ECG signals and heart sounds undetectable to the human ear

  • Trialed on ~12,000 patients across 200 GP practices

  • 2× faster diagnosis of heart failure, 3× faster for atrial fibrillation, nearly 2× faster for valve disease

  • Delivers results in just 15 seconds via cloud-based AI analysis

This breakthrough could bring life-saving diagnoses into GP clinics, enabling earlier treatment and reducing emergency hospital admissions.

➡️ Related: In Issue #14, a Mayo Clinic study showed an AI stethoscope doubled detection of pregnancy-related heart failure in Nigeria, highlighting its global impact. Read more here →

➡️ Also worth noting: Amsterdam-based startup Lapsi Health has developed Keikku, an AI-powered stethoscope that listens at 45 body points, detects 11 pathological sounds, and converts them into digital biomarkers. Already in pilot use across Europe, it highlights how AI stethoscopes are expanding beyond the UK, reshaping diagnostics worldwide. Read our deep dive on Keikku here →

Image generated by Nano Banana

Researchers at UC San Diego developed a new AI segmentation system that improves accuracy while slashing data needs for training:

Source: Nature Communications

  • Generates synthetic training data to reduce annotation needs by up to 20×

  • Boosts segmentation accuracy by 10–20% across 11 medical imaging tasks

  • Tested on 19 datasets, including ultrasound, dermoscopy, colonoscopy, and fetoscopy

  • Works in 2D and 3D scans, mapping organs such as the liver and hippocampus

This approach lowers costs and removes barriers for smaller hospitals, making high-performance imaging AI more accessible worldwide.

Sponsored

Training cutting edge AI? Unlock the data advantage today.

If you’re building or fine-tuning generative AI models, this guide is your shortcut to smarter AI model training. Learn how Shutterstock’s multimodal datasets—grounded in measurable user behavior—can help you reduce legal risk, boost creative diversity, and improve model reliability.

Inside, you’ll uncover why scraped data and aesthetic proxies often fall short—and how to use clustering methods and semantic evaluation to refine your dataset and your outputs. Designed for AI leaders, product teams, and ML engineers, this guide walks through how to identify refinement-worthy data, align with generative preferences, and validate progress with confidence.

Whether you're optimizing alignment, output quality, or time-to-value, this playbook gives you a data advantage. Download the guide and train your models with data built for performance.

Image generated by Nano Banana


South Australia’s Sunrise electronic medical record (EMR) is being enhanced with a new AI-driven system, the CALHN Critical Care Informatics System (CCCIS, or “Kiss”):

  • Integrates directly with Sunrise across all public hospitals

  • Mimics old bedside ICU paper charts for speed and familiarity

  • Tested with doctors and nurses on “workstations on wheels” in busy wards

  • Uses AI to track vital sign trends, IV use, and predict risks like acute kidney injury

Still at the research stage, KISS aims to scale from ICU pilots to statewide use—bringing AI-powered decision support into frontline care while keeping clinicians in control.

Sponsored

The Future of AI in Marketing. Your Shortcut to Smarter, Faster Marketing.

This guide distills 10 AI strategies from industry leaders that are transforming marketing.

  • Learn how HubSpot's engineering team achieved 15-20% productivity gains with AI

  • Learn how AI-driven emails achieved 94% higher conversion rates

  • Discover 7 ways to enhance your marketing strategy with AI.

Image generated by AI

A new machine learning model from Chiang Mai University (Thailand) improves predictions of which emergency patients need ICU admission compared to standard triage systems:

  • Trained on 163,452 ED visits (2018–2022)

  • XGBoost model outperformed CTAS (AUROC 0.917 vs. 0.882)

  • Precision-Recall performance nearly doubled (0.629 vs. 0.333)

  • Key predictors: mode of arrival, age, vital signs, and free-text chief complaints via multilingual embeddings

By integrating structured and unstructured EHR data, the model reduces under-triage and improves resource allocation, offering a scalable path for safer, more consistent ED decision-making.

🩺 AI Takes Aim at Skin Cancer in NHS Trial

England’s National Health Service is piloting DERM, an AI-powered skin-lesion analysis tool designed to triage cancer risk as accurately as physicians:

  • Uses a smartphone + dermoscopic lens to capture lesion images

  • Photos are uploaded to the DERM online platform for AI analysis

  • Flags suspicious lesions for urgent review, reducing delays in care

  • Being trialed across multiple NHS trusts during a 3-year study

  • Concerns remain about effectiveness on darker skin tones and the lack of clinician involvement in triage

If proven effective, DERM could streamline dermatology diagnosis, save costs, and free up specialist time—while raising important questions about equity and oversight.

Sponsored

Start learning AI in 2025

Keeping up with AI is hard – we get it!

That’s why over 1M professionals read Superhuman AI to stay ahead.

  • Get daily AI news, tools, and tutorials

  • Learn new AI skills you can use at work in 3 mins a day

  • Become 10X more productive

🩺 Your AI Co-Pilot in Healthcare

Image generated by Nano Banana

If AI could give you one superpower, which would you pick?

Login or Subscribe to participate in polls.

Stay Ahead: AI Healthcare News!

This is just a glimpse of the AI healthcare revolution. Subscribe for exclusive updates on medicine’s future!