Advancing non-invasive melanoma diagnostics with deep learning and multispectral photoacoustic imaging
(2025) In Photoacoustics 45.- Abstract
The incidence of melanoma is rising and will require more efficient diagnostic procedures to meet a growing demand. Excisional biopsy and histopathology is still the standard, which often requires multiple surgical incisions with increasing margins due inaccurate visual assessment of where the melanoma borders to healthy tissue. This challenge stems, in part, from the inability to reliably delineate the melanoma without visually inspecting chemically stained histopathological cross-sections. Spectroscopic imaging have shown promise to non-invasively characterize the molecular composition of tissue and thereby distinguish melanoma from healthy tissue based on spectral features. In this work we describe a computational framework applied... (More)
The incidence of melanoma is rising and will require more efficient diagnostic procedures to meet a growing demand. Excisional biopsy and histopathology is still the standard, which often requires multiple surgical incisions with increasing margins due inaccurate visual assessment of where the melanoma borders to healthy tissue. This challenge stems, in part, from the inability to reliably delineate the melanoma without visually inspecting chemically stained histopathological cross-sections. Spectroscopic imaging have shown promise to non-invasively characterize the molecular composition of tissue and thereby distinguish melanoma from healthy tissue based on spectral features. In this work we describe a computational framework applied to multispectral photoacoustic (PA) imaging data of melanoma in humans and demonstrate how the borders of the tumor can be automatically determined without human input. The framework combines K-means clustering, for an unbiased selection of training data, a one-dimensional convolutional neural network applied to PA spectra for classifying pixels as either healthy or diseased, and an active contour algorithm to finally delineate the melanoma in 3D. The work stands to impact clinical practice as it can provide both pre-surgical and perioperative guidance to ensure complete tumor removal with minimal surgical incisions.
(Less)
- author
- Merdasa, Aboma
LU
; Fracchia, Alice ; Stridh, Magne LU ; Hult, Jenny LU
; Andersson, Emil LU
; Edén, Patrik LU ; Olariu, Victor LU and Malmsjö, Malin LU
- organization
-
- Ophthalmology, Lund
- LTH Profile Area: Engineering Health
- LU Profile Area: Light and Materials
- LTH Profile Area: Photon Science and Technology
- Ophthalmology Imaging Research Group (research group)
- Centre for Environmental and Climate Science (CEC)
- Computational Science for Health and Environment (research group)
- BECC: Biodiversity and Ecosystem services in a Changing Climate
- MERGE: ModElling the Regional and Global Earth system
- Clinical and experimental lung transplantation (research group)
- publishing date
- 2025-10
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Clinical translation, Deep learning, Melanoma, Photoacoustic imaging, Spectroscopy
- in
- Photoacoustics
- volume
- 45
- article number
- 100743
- publisher
- Elsevier
- external identifiers
-
- scopus:105009741400
- pmid:40686556
- ISSN
- 2213-5979
- DOI
- 10.1016/j.pacs.2025.100743
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2025 The Authors
- id
- f5e4c5e5-00b8-4f52-9b85-2e6e0291be82
- date added to LUP
- 2025-09-12 14:17:09
- date last changed
- 2025-09-26 18:57:32
@article{f5e4c5e5-00b8-4f52-9b85-2e6e0291be82, abstract = {{<p>The incidence of melanoma is rising and will require more efficient diagnostic procedures to meet a growing demand. Excisional biopsy and histopathology is still the standard, which often requires multiple surgical incisions with increasing margins due inaccurate visual assessment of where the melanoma borders to healthy tissue. This challenge stems, in part, from the inability to reliably delineate the melanoma without visually inspecting chemically stained histopathological cross-sections. Spectroscopic imaging have shown promise to non-invasively characterize the molecular composition of tissue and thereby distinguish melanoma from healthy tissue based on spectral features. In this work we describe a computational framework applied to multispectral photoacoustic (PA) imaging data of melanoma in humans and demonstrate how the borders of the tumor can be automatically determined without human input. The framework combines K-means clustering, for an unbiased selection of training data, a one-dimensional convolutional neural network applied to PA spectra for classifying pixels as either healthy or diseased, and an active contour algorithm to finally delineate the melanoma in 3D. The work stands to impact clinical practice as it can provide both pre-surgical and perioperative guidance to ensure complete tumor removal with minimal surgical incisions.</p>}}, author = {{Merdasa, Aboma and Fracchia, Alice and Stridh, Magne and Hult, Jenny and Andersson, Emil and Edén, Patrik and Olariu, Victor and Malmsjö, Malin}}, issn = {{2213-5979}}, keywords = {{Clinical translation; Deep learning; Melanoma; Photoacoustic imaging; Spectroscopy}}, language = {{eng}}, publisher = {{Elsevier}}, series = {{Photoacoustics}}, title = {{Advancing non-invasive melanoma diagnostics with deep learning and multispectral photoacoustic imaging}}, url = {{http://dx.doi.org/10.1016/j.pacs.2025.100743}}, doi = {{10.1016/j.pacs.2025.100743}}, volume = {{45}}, year = {{2025}}, }