Artificial intelligence (AI) and machine learning provide new opportunities for precision medicine, where data-driven approaches are applied for improved diagnostics, prognostication and treatment decisions. One of the medical disciplines that are becoming increasingly data-driven is pathology, where digital scanning of tissue samples is becoming a routine practice and will provide vast amounts of image data usable as the basis for more efficient and accurate clinical management of diseases like cancer.
In our group, we apply the latest AI techniques to the analysis of digital pathology data with the aim of improving the efficiency, accuracy and reproducibility of pathological assessments. By developing AI-based decision support tools, routine tasks can be partially or fully automated, and a tireless “digital colleague” provided to pathologists who struggle with an increasing workload and the demand for more extensive and precise quantification.
Taking a step further, we increasingly work on multi-modal analytics, where image data is processed together with molecular information and clinical variables to build AI models capable of estimating the most likely future course of an individual’s disease and predicting optimal therapeutic options.
The core competences of the group include large-scale image processing and analysis, development of deep learning based AI algorithms, and efficient utilization of high-performance computing systems. We work together with a wide international network of clinical collaborators to reach these aims and to ensure that our AI solutions are applicable across diverse clinical settings and patient populations.
Group members
Sol Erika Boman (PhD student, principal supervisor)
Xiaoyi Ji (PhD student, co-supervisor)
Kelvin Szolnoky (PhD student, co-supervisor)
Andrea Camilloni (PhD student, co-supervisor)