npj Digital Medicine

Our latest journal metrics are now available! nature.com/npjdigitalmed/…

Just as democracy depends on a free and independent press to hold power accountable, medical progress depends on independent journals to vet, challenge and advance science without political interference. washingtonpost.com/opinions/2025/…

Pitfalls of large language models (#LLM) in #medical #ethics reasoning | @npjDigitalMed nature.com/articles/s4174… #MedTwitter #AImedicine #AI #HealthAI #AIinHealthcare #MedTech

🎉👏🏽 Yet another exciting paper led by @UofT_TCAIREM Educational Trainee Affiliate Ben Li (@ben_li123)! Predicting outcomes following endovascular aortoiliac revascularization using #machinelearning (#ML) | @npjDigitalMed nature.com/articles/s4174… #MedTwitter #AImedicine #AI

AI models developed for use in general populations are not uniformly transportable into oncology. Few rigorous studies of large language models have been done in cancer care. nature.com/articles/s4174…

Proprietary large language models demonstrate good structured text conversion for radiology reporting. However, their implementation in healthcare has significant privacy challenges. nature.com/articles/s4174…

💬 The use of LLMs for cancer care has grown rapidly in recent years. But a recent meta-analysis published in @npjDigitalMed, co-authored by #JameelClinic PI @MarzyehGhassemi, found that the avg diagnostic accuracy was 67.4% w/ a lack of human oversight. nature.com/articles/s4174…

Large language models display subtle blind spots in complex reasoning tasks. This paper illustrates these pitfalls in medical ethics reasoning. nature.com/articles/s4174…

🎉 New paper by @UofT_TCAIREM #Education Affiliate Ben Li (@ben_li123)! "Commercialization of #medical #artificialintelligence technologies: challenges and opportunities @npjDigitalMed" nature.com/articles/s4174… #MedTwitter #AImedicine #AI #HealthAI #AIinHealthcare #MedTech

Despite the growing interest in AI for histopathology, few systems have achieved clinical integration, largely due to the absence of end-to-end, multifunctional solutions. These authors are trying to build that solution. nature.com/articles/s4174…

Applications for our News and Views student editor position are due July 25. Apply soon! nature.com/npjdigitalmed/…

💡 Great insights from Li, Powell & @RegentLee in @npjDigitalMed: AI tools like AAA screening show promise, but commercialisation hurdles remain. For ENT, aligning innovation with regulation is key to real-world impact. #AIinHealthcare #ENT 🤖📈 nature.com/articles/s4174…

16 cancer types and 56 papers. Our systematic review focuses on how Large Language Models were integrated in the cancer decision making process. We are very excited to share our new publication at @SpringerNature @npjDigitalMed (doi: nature.com/articles/s4174…)! (1/n)

Will #AI become our Co-PI? | @npjDigitalMed nature.com/articles/s4174… #MedTwitter #AImedicine #AI #HealthAI #AIinHealthcare #MedTech

Digital health tools need user engagement to work well. In this new review/consensus statement, we 1) lack of agreed-upon metrics for engagement, 2) lack of evidence on how better engagement improves outcomes, and 3) lack of standards for user involvement. nature.com/articles/s4174…

What if we did not need so many specialized health apps? What if transdiagnostic apps that can help across many conditions worked just as well? New review led by @j_linardon has the answer-transdiagnostic as good as specialized apps! Free in @npjDigitalMed nature.com/articles/s4174…

Online platforms fill the needs of patients by offering peer-to-peer support, information on latest studies of a particular condition and ‘lived experiences’, which are often unfulfilled by traditional health care services. nature.com/articles/s4174…

Transdiagnostic-focused apps produce comparable effect sizes to disorder-specific apps — our new meta in @npjDigitalMed

This study underscores the limitations of frontier large language models for up-to-date medical use cases. ai.nejm.org/doi/full/10.10…

Treatment programs delivered through technological means offer a promising solution to increase the availability and scalability of evidence-based psychological interventions. nature.com/articles/s4174…

This perspective examines how large language models can reshape the laboratory model as de facto “Co-PIs” for tasks ranging from literature triage to hypothesis generation. nature.com/articles/s4174…