A study led by Karolinska Institutet in Sweden has demonstrated that the use of an AI tool can improve the pathologists’ examinations of tissue samples obtained from skin cancer tumours, showing accuracy in patients’ prognoses.
Conducted in partnership with Yale University, the research focused on the role of tumour-infiltrating lymphocytes (TILs), which are a critical biomarker in various cancers, including malignant melanoma, also known as skin cancer.
These lymphocytes are immune cells present in or around the tumour and are indicative of the body’s response to cancer.
Researchers investigated the way in which pathological evaluations were affected by an AI tool that is trained for TIL quantification.
Karolinska Institutet鈥檚 study involved 98 pathologists and professionals from related fields, divided into two groups.
The first group comprised experienced pathologists who assessed the TIL amount in digital images of stained tissue sections according to present guidelines.
US Tariffs are shifting - will you react or anticipate?
Don鈥檛 let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataThe second group, consisting of pathologists and researchers from other professions with pathological image assessment expertise, received AI support to quantify TILs.
Everyone evaluated 60 tissue sections from individuals with malignant melanoma in the retrospective study, meaning the samples came from patients whose diagnoses and treatments were already established.
The AI-supported assessments showed significantly higher reproducibility.
The Swedish Society for Medical Research and Region Stockholm provided the funding for the research. It also secured various grants from the US National Institutes of Health.
Karolinska Institutet department of oncology-pathology聽associate professor Balazs Acs said: “Understanding the severity of a patient’s disease based on tissue samples is important, among other things, for determining how aggressively it should be treated. We now have an AI-based tool that can quantify the TIL biomarker, which could help with treatment decisions in the future.
鈥淗owever, more studies are needed before this AI tool can be used in clinical practice, but the results so far are promising and suggest that it could be a very useful tool in clinical pathology.鈥�
