Facilitating clinically relevant skin tumor diagnostics with spectroscopy-driven machine learning

• Published in: iScience Vol. 27; no. 5; p. 109653
• Main Authors: Andersson, Emil, Hult, Jenny, Troein, Carl, Stridh, Magne, Sjögren, Benjamin, Pekar-Lukacs, Agnes, Hernandez-Palacios, Julio, Edén, Patrik, Persson, Bertil, Olariu, Victor, Malmsjö, Malin, Merdasa, Aboma
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 17.05.2024
• Subjects: Biofysik; Biophysics; Cancer and Oncology; Cancer och onkologi; Clinical Medicine; Computer science; Fysik; Health sciences; Klinisk medicin; Medical and Health Sciences; Medicin och hälsovetenskap; Natural Sciences; Naturvetenskap; Physical Sciences; Computer science; Natural sciences; Health sciences
• ISSN: 2589-0042; 2589-0042
• DOI: 10.1016/j.isci.2024.109653

In the dawning era of artificial intelligence (AI), health care stands to undergo a significant transformation with the increasing digitalization of patient data. Digital imaging, in particular, will serve as an important platform for AI to aid decision making and diagnostics. A growing number of studies demonstrate the potential of automatic pre-surgical skin tumor delineation, which could have tremendous impact on clinical practice. However, current methods rely on having ground truth images in which tumor borders are already identified, which is not clinically possible. We report a novel approach where hyperspectral images provide spectra from small regions representing healthy tissue and tumor, which are used to generate prediction maps using artificial neural networks (ANNs), after which a segmentation algorithm automatically identifies the tumor borders. This circumvents the need for ground truth images, since an ANN model is trained with data from each individual patient, representing a more clinically relevant approach. [Display omitted] •Hyperspectral imaging enables noninvasive skin tumor characterization•Novel AI approach bypasses need for ground truth images in tumor delineation•Individual ANN models per patient enhance clinical relevance and objectivity•Promising for patient adaptability in future differential skin tumor diagnostics Health sciences; Natural sciences; Computer science