It’s widely assumed that drug treatment is a miracle of modern science. However, it’s equally arguable that modern drug treatment is a very expensive disaster, with serious consequences for public health. For example, a drug typically takes 10-12 years to develop and the cost is as high as $2.5 billion. And yet despite that, 90% of drug candidates fail to gain approval. And, especially for chronic diseases, their overall benefit is often poor or even non-existent.

Can AI help?

With Dr Scott Chetham

Episode 53: Radiology AI: Roadblocks and how to get past them. With Professor Paul Chang.

A recent study by McKinney and colleagues showed that deep learning AI systems outperformed human radiologists for detection of breast cancer.

Except that this wasn't so recent... it was in fact published in January 2020, over 5 years ago. While AI screening is growing it still only accounts for around 10-20% ofroutine mammograms. So what’s the hold-up in adoption? And how do we get past the roadblocks?

The Medical AI Podcast is a top-rated podcast on FeedSpot. Check out other FeedSpot podcasts here:⁠https://podcast.feedspot.com/uk_ai_podcasts/⁠

A recent study by McKinney and colleagues showed that deep learning AI systems outperformed human radiologists for detection of breast cancer. Except that this wasn't so recent. it was in fact published in January 2020, over 5 years ago. While AI screening is growing it still only accounts for around 10-20% of routine mammograms. So what’s the hold-up in adoption? And how do we get past the road blocks?

One of the greatest health challenges for western countries is obesity. Why is that an issue? Obesity increases the risk of all major diseases including heart disease, cancer, dementia, and diabetes. What role is AI playing in addressing this? With Dr Sera Lavelle, Clinical Psychologist.

The Medical AI Podcast is a top-rated podcast on FeedSpot. Check out other FeedSpot podcasts here: ⁠https://podcast.feedspot.com/uk_ai_podcasts/⁠ 

One of the greatest health challenges for western countries is obesity. Why is that an issue? Obesity increases the risk of all major diseases including heart disease, cancer, dementia, and diabetes. What role is AI playing in addressing this? With Dr Sera Lavelle, Clinical Psychologist.

The Medical AI Podcast is a top-rated podcast on FeedSpot. Check out other FeedSpot podcasts here: ⁠https://podcast.feedspot.com/uk_ai_podcasts/⁠ 

Radiology could be the most important area of medical AI. Most FDA approvals in medical AI are for radiology – with almost 1000 approvals since 2017. So radiology is undergoing a major AI transformation. But the abundance of radiology AI tools has created a big problem. With so much choice, healthcare providers are often confused about

◦ which tools to deploy

◦ how to deploy them

◦ how they will work

◦ whether they could conflict with current computer systems

This uncertainty is a significant brake on progress. A part of the solution to that problem is that of AI marketplaces, or platforms. And what are the lessons for other areas of medical AI? With me to discuss this important area is Pelu Tran, CEO of Ferrum Health

Episode 50: AI and the non-invasive "brain biopsy" (1H-MRS) with Professor Alex Lin of Harvard Medical School

Imagine a doctor telling you you need a biopsy. Then he or she tells you the biopsy is going to be from your brain. You might well panic. But then you learn that there will be no drilling through the skull. This is a non-invasive biopsy. It uses a technique known as H-MRS and it could be the coolest type of neuroimaging you have never heard of.

So what is it? How can it help? And how is AI advancing this technique?

With Professor Alex Lin of Harvard Medical School.

The classic American TV series Six Million Dollar Man followed the story of Steve Austin, who had to be rebuilt after near-fatal injuries in a plane crash. He was Better than he was before… Better...stronger...faster." Prosthetic limbs have been helping people for many years even centuries but what is the state-of-the-art? How is it helping amputees? And how far can the technology take us towards a science fiction future? With Blair Lock, CEO of Coapt Engineering.

The Medical AI Podcast is a top-rated podcast on FeedSpot. Check out other FeedSpot podcasts here: ⁠https://podcast.feedspot.com/uk_ai_podcasts/⁠ 

The classic American TV series Six Million Dollar Man followed the story of Steve Austin, who had to be rebuilt after near-fatal injuries in a plane crash. He was Better than he was before… Better...stronger...faster." Prosthetic limbs have been helping people for many years even centuries but what is the state-of-the-art? How is it helping amputees? And how far can the technology take us towards a science fiction future? With Blair Lock, CEO of Coapt Engineering.

The Medical AI Podcast is a top-rated podcast on FeedSpot. Check out other FeedSpot podcasts here: ⁠https://podcast.feedspot.com/uk_ai_podcasts/⁠ 

Skin cancer is one of the most common forms of cancer globally, with millions of cases diagnosed each year. Its prevalence is rising due to factors like increased sun exposure, aging populations, and lifestyle changes. The main types of skin cancer include melanoma, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC), with melanoma being the deadliest. Early detection is crucial for successful treatment, as skin cancer can often be treated effectively if caught early.

However, diagnosing skin cancer can be challenging due to the wide variety of skin conditions and the complexity of distinguishing between benign and malignant growths. Traditional methods of detection, such as visual inspection by dermatologists and biopsies, are time-consuming, subjective, and prone to human error.

AI systems, particularly deep learning and computer vision models, are being increasingly applied to help address these challenges. These AI tools can analyze medical images of skin lesions with high accuracy, sometimes surpassing dermatologists in diagnosing skin cancer. They can quickly identify suspicious moles or growths, enabling earlier and more accurate diagnosis. AI systems can also assist in classifying lesions into categories (benign or malignant) and track changes over time, improving long-term monitoring.

The potential benefits of AI in skin cancer detection include faster diagnosis, reduced workload for healthcare professionals, and improved patient outcomes through early intervention. As AI technology continues to evolve, it may become a key tool in reducing the global burden of skin cancer.

With Cody Simmons, CEO of Dermasensor

Check out other top-rated AI podcasts on FeedSpot: