Scientists at Wayne State University are developing artificial intelligence models to reduce the cost of complex drug designs, while a Stanford study found that ChatGPT-4 achieved 92% medical diagnostic accuracy. , and outperformed 74% of human doctors. This wave of AI breakthroughs is also reflected in financial markets, with health tech stocks up 12% in 2024, AI healthcare companies now valued at up to five times that of non-AI companies, and venture capital attracts 38% of healthcare investment.
AI models could reduce drug development costs
The Wayne State University project aims to leverage AI and computer models to design new drugs, with a primary focus on complex classes of drugs that have traditionally been difficult to simulate.
“Most drug design is done using small organic molecules,” Wayne State University assistant professor of chemistry Alice Walker said in a news release Monday (Oct. 28). “The computational models we need to do this are excellent, but scaffolding exotic drugs such as sugars and fluorescent molecules is problematic. Our lab will use this grant to We plan to develop new computational techniques and apply them to the development of new carbohydrate drugs and fluorescent imaging agents.”
The research, funded by the National Institutes of Health, could lead to breakthrough treatments with fewer side effects. Walker’s team will combine simulation and machine learning to better understand the behavior of molecules, particularly in the development of cancer drug targets.
AI outperforms doctors in medical diagnostic research
A new study from Stanford University reveals that ChatGPT-4 achieved high diagnostic accuracy, scoring 92% when analyzing real patient cases. This significantly exceeded the scores of traditional and AI-assisted physicians (74% and 76%, respectively).
Physicians with access to ChatGPT did not see significant improvements in diagnostic reasoning, but completed assessments slightly faster compared to those without AI assistance. The findings suggest that healthcare professionals may need better training to effectively utilize AI tools in clinical practice.
“Our study shows that ChatGPT has the potential to be a powerful tool in medical diagnostics. Therefore, we are disappointed that physicians’ use of ChatGPT did not significantly improve their clinical reasoning. “We were surprised,” said study co-lead author Ethan Goh, a postdoctoral fellow at Stanford University School of Medicine. he said in a blog post.
The study, published in JAMA Network Open, emphasizes that while AI will not replace doctors, it has the potential to improve their abilities. In response to these findings, several leading healthcare organizations launched ARiSE, a bicoastal network dedicated to evaluating AI applications in healthcare. This study highlights the need to improve physician-AI collaboration and properly integrate AI tools in clinical practice.
Health tech sector recovers
Health tech stocks rose 12% in 2024, with private market valuations returning to peak levels, driven largely by investments in artificial intelligence, according to Bessemer Venture Partners’ annual industry report. The sector’s public market capitalization now stands at $98 billion, with 34 companies including Tempus and Waystar, which held initial public offerings (IPOs) earlier this year.
In their 2024 State of Health Tech Report, Bessemer partners Sofia Guerra and Steve Kraus say the sector is “demonstrating remarkable adaptability and strength in the face of ongoing market challenges.” “I’m doing it,” he wrote.
AI-focused companies currently attract 38% of venture capital healthcare funding, with valuations two to five times higher than non-AI companies. However, early-stage funding remains a challenge, with Series A companies taking 50% longer to raise capital than in previous years.
More information: AI, Artificial Intelligence, ChatGPT, Digital Transformation, Drug Discovery, Fundraising, Medical Technology, Healthcare, Investments, Medical Research, National Institutes of Health, News, Pharmaceuticals, PYMNTS News, Stanford University, Technology, Wayne State University
Source link
