• Supermedic.ai
  • Posts
  • 🔬 AI spots cancer and viral infections with nanoscale precision

🔬 AI spots cancer and viral infections with nanoscale precision

ALSO: 50% of FDA-approved AI medical devices are not trained on real patient data, Neurii project: using AI to analyze 1.6 million brain scans to boost dementia prediction, This AI-powered sensor can 'smell' disease

In partnership with

Hey!

Welcome to this week’s edition of Supermedic, where we explore the latest developments in artificial intelligence and its transformative impact on healthcare.

Let’s dive in!

Victor

TODAY’S MENU

  • AI Spots Cancer and Viral Infections at Nanoscale Precision

  • Almost 50% of FDA-approved AI Medical Devices Are Not Trained on Real Patient Data

  • NEURii Project: Using AI to Analyze 1.6 Million Brain Scans to Boost Dementia Prediction

  • This AI-Powered Sensor Can 'Smell' Disease

  • Everything Else You Should Know this Week

    Read time: under 6 minutes

RESEARCH

The image uses two colors to show specific nuclear components which allow researchers to see detailed structures within the cell nucleus at nanoscale resolution. Credit: Zhong Limei

Researchers have developed an AI that can distinguish cancer cells from normal ones and detect early viral infections by analyzing cellular changes as small as 20 nanometers—far beyond traditional methods.

How It Works

  • AINU uses advanced AI (convolutional neural network) to analyze highly detailed images taken from STORM microscope, revealing tiny changes that traditional methods might miss.

  • The AI detects structural changes as small as 20 nanometers—5,000 times thinner than a human hair.

  • It detected signs of a herpes virus infection just one hour after the infection started by noticing small changes in the cell’s DNA.

  • AINU can also accurately identify stem cells, which could speed up research and reduce the need for animal testing.

Why it matters: With AI tools like AINU emerging rapidly, we’re on the brink of a revolution in early disease detection and scientific research. While clinical use might still be on the horizon, the potential impact on healthcare is immense.

OUR SPONSOR

Effective hair loss solutions doctors trust

Hims knows the hair loss struggle is real. That’s why they offer access to a variety of hair loss treatments for men with doctor-trusted ingredients (and have the 5-star reviews to back them up!).

Dealing with a receding hairline, bald spots, or all-over thinning? Hims has a solution that can help you start regrowing your hair in just 3-6 months**.

Start your hair regrowth journey today. After completing an intake form, a licensed medical provider will review and determine if a prescription is appropriate. If prescribed, your medication ships for free!

No doctor visits, no boring waiting rooms, and no never-ending pharmacy lines.

Try their hair loss treatments for men today (and save when you subscribe).***

*Hair Hybrids are compounded products and have not been approved by the FDA. The FDA does not verify the safety or effectiveness of compounded drugs. Restrictions apply, see website for full details. Prescription products require an online consultation with a healthcare provider who will determine if a prescription is appropriate. **Based on separate individual studies of oral minoxidil and oral finasteride. ***Actual price to customer will depend on product and subscription plan purchased.

REGULATION

Illustration by Midjourney

A recent study led by researchers from multiple institutions, including UNC and Duke University, reveals significant gaps in the clinical validation of AI medical devices authorized by the FDA. The findings, published in Nature Medicine, highlight the need for stricter standards in evaluating AI devices used in healthcare.

Study Details:

  • Of the 521 AI medical devices authorized by the FDA, 43% lacked clinical validation data, raising concerns about their reliability.

  • Devices were validated as follows: 144 retrospectively with past data, 148 prospectively with real-time patient data, and only 22 through randomized controlled trials, the gold standard in medical research.

  • Researchers urge the FDA to establish more stringent guidelines that clearly differentiate between validation methods

Why It Matters: With the rapid increase in AI device authorizations—rising from 2 to 69 per year since 2016—there is an urgent need for transparent and rigorous validation processes to ensure patient safety and trust in these technologies.

NEUROSCIENCE

Credit: Getty

Scientists from the Universities of Edinburgh and Dundee are launching a massive AI-driven study of over 1.6 million brain scans to develop tools for early dementia prediction and diagnosis.

The details:

  • The project, called NEURii, will use AI and machine learning to analyze CT and MRI scans from Scottish patients over the past decade.

  • The goal is to create advanced digital tools that will help radiologists assess dementia risk during routine brain scans, making early detection more accessible.

  • By matching brain scan data with detailed health records, researchers aim to identify subtle patterns linked to dementia, which could lead to earlier and more accurate diagnosis.

Why it matters: With dementia cases projected to soar to 153 million globally by 2050, the impact of this research could be profound, potentially enabling early intervention and more effective treatment strategies.

SEEN ON INSTAGRAM

Indian Clinic Introduces AI Dental Check-Up

Prime Dental Clinic in India now provides free AI-powered dental check-ups using scanO AI, a device that can swiftly and contactlessly identify the five most common dental issues in under a minute.

AI SAVES LIVES

Credit: Noze

Canadian startup Noze has developed DiagNoze, a portable device that leverages AI to diagnose diseases through breath analysis. This innovative tool could revolutionize medical diagnostics, particularly in underprivileged regions where access to healthcare is limited.

Key Details:

  • DiagNoze detects volatile organic compounds (VOCs) in exhaled breath, key indicators of diseases like tuberculosis.

  • The device offers quick and affordable diagnostics without the need for complex lab equipment, making it a game-changer for healthcare in low-resource areas.

  • Supported by a $1.8 million grant from the Bill and Melinda Gates Foundation, DiagNoze is set to undergo clinical trials in regions with high tuberculosis rates, aiming to curb this deadly disease.

Why It Matters: DiagNoze could transform healthcare by enabling preventive diagnostics, allowing doctors to detect and treat diseases before symptoms even appear. This shift from reactive to proactive healthcare could save lives and reduce the burden on healthcare systems worldwide.

Recommended Newsletter

Must-Read AI Healthcare News This Week

Drug Discovery: ActFound, an AI model developed by American and Chinese scientists, outperformed nine models in predicting cancer drug bioactivity, addressing small dataset issues and potentially accelerating drug development.

Neuroscience: EPFL develops a miniaturized brain-machine interface (MiBMI) that converts brain activity into text with 91% accuracy, aiming to enhance communication for patients with severe motor impairments like ALS.

Biomedical Engineering: University of Waterloo develops a 3D-printable biopolymer that mimics bone tissue, offering a custom solution for skeletal repair, potentially replacing metal implants and improving graft success.

Radiology: University of Toronto researchers used GPT-4 to analyze FDA data on interventional radiology, achieving 86.4% accuracy in classifying safety events.

Mental Health: Slingshot AI raised $30 million, led by Andreessen Horowitz, boosting its valuation to $220 million. The startup’s AI therapist, Ash, addresses mental healthcare shortages with 24/7 interactive counseling.

Oncology: University of Cologne researchers developed an AI-based platform for fast, accurate analysis of lung cancer tissue, automating tumor identification and enhancing diagnostic precision.

Cardiology: Chinese researchers developed an AI model using tongue imaging and risk factors to diagnose coronary artery disease (CAD) with over 99% accuracy.

Thanks for reading!

How would you rate today's newsletter?

Your feedback helps me create better emails for you!

Login or Subscribe to participate in polls.

Reply

or to participate.