Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A spectroscopy-based machine learning framework on photoacoustic imaging data for automatic 3D delineation of melanoma tumors

Merdasa, Aboma LU orcid ; Fracchia, Alice ; Stridh, Magne LU ; Hult, Jenny LU orcid ; Andersson, Emil LU orcid ; Edén, Patrik LU ; Olariu, Victor LU and Malmsjö, Malin LU orcid (2025) Photons Plus Ultrasound: Imaging and Sensing 2025 In Progress in Biomedical Optics and Imaging - Proceedings of SPIE 13319.
Abstract

The incidence of melanoma is rising, increasing the need for efficient diagnostics. Excisional biopsy and histopathology remain the “gold standard” but often require multiple surgical incisions due to limitations in making a visual assessment of melanoma borders either by eye or using dermatoscopy. This challenge partly arises from the inability to non-invasively produce the necessary spectral contrast to reliably delineate melanomas without chemically staining the tissue. Spectroscopic imaging provides an intriguing alternative to non-invasively characterize the molecular composition of tissue, and on that basis distinguish melanoma from healthy tissue. This work presents a computational framework for multispectral photoacoustic (PA)... (More)

The incidence of melanoma is rising, increasing the need for efficient diagnostics. Excisional biopsy and histopathology remain the “gold standard” but often require multiple surgical incisions due to limitations in making a visual assessment of melanoma borders either by eye or using dermatoscopy. This challenge partly arises from the inability to non-invasively produce the necessary spectral contrast to reliably delineate melanomas without chemically staining the tissue. Spectroscopic imaging provides an intriguing alternative to non-invasively characterize the molecular composition of tissue, and on that basis distinguish melanoma from healthy tissue. This work presents a computational framework for multispectral photoacoustic (PA) imaging data of melanoma skin tumors that automatically determines tumor borders without human input. The framework combines K-means clustering for training data generation, a 1D convolutional neural network to classify pixels based on their spectral features, and an active contour algorithm to delineate the tumor in both 2D and 3D. An important feature of our model is that the training data is contained within one patient and does not rely on a population-based signature, yielding an individualized diagnostic approach that is of high relevance for precision skin tumor diagnostics. Non-invasive determination of melanoma tumor borders can improve clinical practice by providing pre-surgical and perioperative guidance for complete tumor removal with minimal incisions.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Breslow’s depth, deep learning, melanoma, non-invasive imaging, Photoacoustic imaging, spectroscopy
host publication
Photons Plus Ultrasound : Imaging and Sensing 2025 - Imaging and Sensing 2025
series title
Progress in Biomedical Optics and Imaging - Proceedings of SPIE
editor
Oraevsky, Alexander A. and Wang, Lihong V.
volume
13319
article number
133190I
publisher
SPIE
conference name
Photons Plus Ultrasound: Imaging and Sensing 2025
conference location
San Francisco, United States
conference dates
2025-01-26 - 2025-01-29
external identifiers
  • scopus:105004319791
ISSN
1605-7422
ISBN
9781510683860
DOI
10.1117/12.3043099
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2025 SPIE.
id
7024d2ac-9f35-41a4-9c0a-596e7f0e621c
date added to LUP
2025-08-19 11:30:57
date last changed
2025-08-19 11:34:13
@inproceedings{7024d2ac-9f35-41a4-9c0a-596e7f0e621c,
  abstract     = {{<p>The incidence of melanoma is rising, increasing the need for efficient diagnostics. Excisional biopsy and histopathology remain the “gold standard” but often require multiple surgical incisions due to limitations in making a visual assessment of melanoma borders either by eye or using dermatoscopy. This challenge partly arises from the inability to non-invasively produce the necessary spectral contrast to reliably delineate melanomas without chemically staining the tissue. Spectroscopic imaging provides an intriguing alternative to non-invasively characterize the molecular composition of tissue, and on that basis distinguish melanoma from healthy tissue. This work presents a computational framework for multispectral photoacoustic (PA) imaging data of melanoma skin tumors that automatically determines tumor borders without human input. The framework combines K-means clustering for training data generation, a 1D convolutional neural network to classify pixels based on their spectral features, and an active contour algorithm to delineate the tumor in both 2D and 3D. An important feature of our model is that the training data is contained within one patient and does not rely on a population-based signature, yielding an individualized diagnostic approach that is of high relevance for precision skin tumor diagnostics. Non-invasive determination of melanoma tumor borders can improve clinical practice by providing pre-surgical and perioperative guidance for complete tumor removal with minimal 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}},
  booktitle    = {{Photons Plus Ultrasound : Imaging and Sensing 2025}},
  editor       = {{Oraevsky, Alexander A. and Wang, Lihong V.}},
  isbn         = {{9781510683860}},
  issn         = {{1605-7422}},
  keywords     = {{Breslow’s depth; deep learning; melanoma; non-invasive imaging; Photoacoustic imaging; spectroscopy}},
  language     = {{eng}},
  publisher    = {{SPIE}},
  series       = {{Progress in Biomedical Optics and Imaging - Proceedings of SPIE}},
  title        = {{A spectroscopy-based machine learning framework on photoacoustic imaging data for automatic 3D delineation of melanoma tumors}},
  url          = {{http://dx.doi.org/10.1117/12.3043099}},
  doi          = {{10.1117/12.3043099}},
  volume       = {{13319}},
  year         = {{2025}},
}