Publication details
AI in cancer research: from histology to personalized medicine
| Authors | |
|---|---|
| Year of publication | 2023 |
| MU Faculty or unit | |
| Citation | |
| Description | Artificial intelligence (AI) is becoming an inevitable part of a modern world medicine and biomedical industry that is becoming more and more digital. AI can analyze huge amounts of data, uncovering invisible patterns, speeding up processes and improving performance in many areas. Moreover, thanks to the process of continuous learning and thanks to updates, AI has an amazing capacity for continuous improvement. In histopathology, AI methods are mainly used in diagnosing diseases and designing treatment plans. They can be used, for example, for automatic classification in diagnosis, prediction of patient survival and response to treatment, segmentation, object detection, and analysis of microscope images[1]. AI methods are often used as a helpful screening tool, predicting the presence or absence of cancer. For example, Wang et al. developed a model that can discriminate between patients with pancreatic ductal adenocarcinoma (PDAC) and healthy controls with 86.74% accuracy[2]. Sohn et al. developed a different AI algorithm that can classify these patients with accuracy 91.00%[3]. Lee et al. used AI to create a model for predicting postoperative survival in PDAC patients. The AUC of the model for predicting 2-year overall survival in the test dataset was 0.76 and 1-year recurrence-free survival 0.74[4]. These results show that the use of AI in histology is promising and can significantly improve the diagnosis and prediction of patient survival. Supported by Ministry of Health of the Czech Republic, grant nr. NU23-08-00241 and by Masaryk University, project nr. MUNI/A/1301/2022. [1] S. Försch, F. Klauschen, P. Hufnagl, and W. Roth, “Artificial Intelligence in Pathology,” Dtsch Arztebl Int, vol. 118, no. 12, pp. 194–204, Mar. 2021, doi: 10.3238/arztebl.m2021.0011. [2] G. Wang et al., “Metabolic detection and systems analyses of pancreatic ductal adenocarcinoma through machine learning, lipidomics, and multi-omics,” Sci Adv, vol. 7, no. 52, p. eabh2724, doi: 10.1126/sciadv.abh2724. [3] A. Sohn et al., “A deep learning model to triage and predict adenocarcinoma on pancreas cytology whole slide imaging,” Sci Rep, vol. 13, p. 16517, Oct. 2023, doi: 10.1038/s41598-023-42045-w. [4] W. Lee et al., “Preoperative data-based deep learning model for predicting postoperative survival in pancreatic cancer patients,” Int J Surg, vol. 105, p. 106851, Sep. 2022, doi: 10.1016/j.ijsu.2022.106851. |
| Related projects: |