ChatGPT Scored 72% in Clinical Decision Accuracy  – This And More News In Digital Health This Week

ChatGPT Scored 72% in Clinical Decision Accuracy – This And More News In Digital Health This Week

A recent study revealing ChatGPT's high scoring in clinical decision accuracy has made waves. While the results are certainly promising, it's crucial to keep in mind that these tests capture only a fraction of the myriad skills and knowledge required to treat patients - and we need to interpret the implications accordingly.

We've rolled out a new YouTube video inviting ideas for future topics that meld the realms of medical futurism, digital health, and AI. Dive in, drop your thoughts in the comments under the video, and don't forget to upvote the suggestions that resonate with you.

CHATGPT SCORED 72% IN CLINICAL DECISION ACCURACY, MGB STUDY SHOWS?????

"ChatGPT proved best in making a final diagnosis, where the ai had 77% accuracy in the study, funded in part by the National Institute of General Medical Sciences.

It was lowest-performing in making differential diagnoses, where it was only 60% accurate, and in clinical management decisions, underperforming at 68% accuracy based on the clinical data the llm was trained on. "

READ MORE


SEE A HOSPITAL FROM THE PERSPECTIVE OF AN ARTIFICIAL INTELLIGENCE

Existing AI systems that help monitor patients could aid in developing algorithms that can provide real-time healthcare recommendations.

There are two sides to this coin: the immense potential these technologies present, and the mountainous challenges related to privacy and ethics that they introduce and which need to be addressed.

READ MORE?


OUR NEW EBOOK: GENERATIVE AI IN HEALTHCARE - THE CHATGPT REVOLUTION

An essential read for anyone aiming to grasp the burgeoning role of AI in the medical sphere. This practical guide demystifies the buzz surrounding generative AI, offering readers a clear, non-technical understanding of its immense potential.

More than just a primer, the ebook assists readers in evaluating their readiness for the new era of AI in medicine. Dive in to uncover the realities and rewards of this technological revolution.

GET IT HERE


WHAT'S THE BEST IDEA? VOTE ON YOUR FAVORITE TOPICS!

What are the best / most important topics in the realm of medical futurism we failed to discuss in the past decade?

We are looking for your input here, so if you have ideas, please come over to YouTube and weigh in under the video, and/or vote for the best suggestions others requested.

HAVE YOUR SAY


CE-MARKED MEDICAL DEVICE AI SUPPORTING DOCTORS IN ACTION

Tonic App is a mobile platform that aggregates resources useful for healthcare professionals to diagnose and treat patients. According to the company, over 120k doctors use it. Its main features include functions like certified calculators and scales for drug dosing and info on drug interactions, discussion of clinical cases and clinical decision trees, healthcare news and jobs.

We published a short interview with them discussing questions like how to make AI safe for medical use.

READ MORE


GOOGLE EXPANDS GENERATIVE AI MODEL MED-PALM TO MORE HEALTH CUSTOMERS

Google's medical large language model trained on medical information has been accessible for Google Cloud customers in the healthcare and life sciences, such as HCA Healthcare, Mayo Clinic and Meditech. Now Google is expanding this access to Med-PaLM to more customers.

Do you have access to Med-PaLM through your company/clinic? I'd be delighted if you reached out to me to share what you think about it and what are your first experiences like.

READ MORE


FIRST IMMUNOTHERAPY DRUG DESIGNED USING GENERATIVE AI TECHNOLOGY

It was just claimed that a company successfully used generative AI to design a new immunotherapy drug for cancer.

The new therapy is designed to target a molecule that is present in many cancers with poor survival rates. In just 11 months, the company could identify a potential drug that binds to receptors with greater affinity and doesn't bind to healthy cells.

READ MORE

MORE NEWS ABOUT THE FUTURE OF MEDICINE

DNA CAGES? –? 13 Technologies That Will Shape The Future Of Cancer Care

OH, YES! – Will It Be A Luxury To Meet Physicians?

METABOLISM DATA – AI-Powered Triage Platform Could Aid Future Viral Outbreak Response

MINIMALLY INVASIVE – In global first, Shaare Zedek spine surgeon combines augmented reality with robotics

Clifton Veach

Founder/Director of Product Design & Strategic Markets - Deaf-Tek Studio

1 年

I am not a Medical Professional and although I have had my share of interactions, I am confused about a Decision Making Success Number between 60-70% being rated as "wonderful" or even at my take of "Not Bad", but that leaves open the question of Personal Diagnostic Accuracy of current physicians... does this mean the current accuracy rate of accurate diagnosis is LESS than 60-70% across the Board??? If that is true, I am glad my Doctors are not folding or loading parachutes for a living, or we would be picking up 30% of jumpers as dead bodies on the ground... The accuracy herein is not clear on external input of testing results, previous or co-existing conditions, serious vs common ailments, etc, so I will withhold my opinion for the time being, but will ask every Doctor I visit if they use AI assistants for diagnostics, and if so, will take my chances with experienced guesswork until more data is in...

Moamel Hameed

maintenance engineering

1 年

Good article

Great article! When done right, A.I. will support clinical reliability/confidence levels.

要查看或添加评论,请登录

Bertalan Meskó, MD, PhD的更多文章

社区洞察

其他会员也浏览了