Dual Psi-Pyramid-based Multi-Scale Encoder-Decoder Network with Minkowski Similarity for Melanoma Skin Lesion Segmentation

DOI

10.2991/978-94-6239-756-9_10How to use a DOI?

Keywords

Preprocessing; Hair removal; Segmentation; Fusion; Minkowski Similarity; Skin lesion detection

Abstract

Melanoma is a difficult type of skin cancer that needs to be identified early in order to increase survival rates. In order to efficiently fight this health issue, it is essential that medical imaging technology be used for timely identification. Melanoma-related mortality can be considerably decreased with early management. This paper aims to use datasets of skin images that are first gathered, providing varied sizes for reliable training and testing. Through preprocessing, the Adaptive Wiener Filter (AWF) efficiently reduces noise while improving image quality. In order to avoid obstructing the lesion boundaries, ZF-Net is then used to remove hair from the images. Then the proposed Psi-Pyramid-based Multi-scale Encoder-Decoder Network (PMED-Net) is then used for skin lesion segmentation, using advanced Convolutional Neural Network (CNN) designs for increased accuracy. The outputs from both segmentation methods are then combined using Minkowski similarity, enhancing overall segmentation accuracy to 98.56%, recall to 99.42%, precision to 98.82%, and F1 score to 99.86%. Better analysis and diagnosis are made possible by the binary images that clearly identify the segmented skin lesions. The proposed Psi-PMED Net aims to enhance the early identification and management of skin lesions, thereby improving imaging methods and ultimately leading to better patient outcomes.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Nitalaksheswara Rao Kolukula
AU  - Jayasree Pinajala
AU  - James Stephen Meka
AU  - Pavan Satish Chandaka
PY  - 2026
DA  - 2026/08/31
TI  - Dual Psi-Pyramid-based Multi-Scale Encoder-Decoder Network with Minkowski Similarity for Melanoma Skin Lesion Segmentation
BT  - Proceedings of the Conference on Bioengineering for Global Health (SYMRESEARCH 2.0 2025)
PB  - Atlantis Press
SP  - 121
EP  - 134
SN  - 2468-5747
UR  - https://doi.org/10.2991/978-94-6239-756-9_10
DO  - 10.2991/978-94-6239-756-9_10
ID  - Kolukula2026
ER  -