Advanced search ?

Object

Enhancing multi-class prediction of skin lesions with feature importance assessment
Download

Title: Enhancing multi-class prediction of skin lesions with feature importance assessment

Creator:

Paulauskaite-Taraseviciene, Agne ; Sutiene, Kristina ; Dimsa, Nojus ; Valiukeviciene, Skaidra

Date:

2024

Resource Type:

artykuł

Contributor:

Woźniak, Marcin - ed. ; Kumar, Yogesh - ed. ; Ijaz, Muhammad Fazal - ed.

Subtitle:

.

Group publication title:

AMCS, volume 34 (2024)

Abstract:

Numerous image processing techniques have been developed for the identification of various types of skin lesions. In real-world scenarios, the specific lesion type is often unknown in advance, leading to a multi-class prediction challenge. The available evidence underscores the importance of employing a comprehensive array of diverse features and subsequently identifying the most important ones as a crucial step in visual diagnostics. For this purpose, we addressed both binary and five-class classification tasks using a small dataset, with skin lesions prevalent in Lithuania. The model was trained using a rich set of 662 features, encompassing both conventional image features and graph-based ones, which were obtained from the superpixel graph generated using Delaunay triangulation. We explored the influence of feature importance determined by SHAP values, resulting in a weighted F1-score of 92.48% for the two-class classification and 71.21% for the five-class prediction.

Publisher:

Zielona Góra: Uniwersytet Zielonogórski

Resource Identifier:

oai:zbc.uz.zgora.pl:87179

DOI:

10.61822/amcs-2024-0041

Pages:

617-629

Source:

AMCS, volume 34, number 4 (2024) ; click here to follow the link

Language:

eng

License CC BY 4.0:

click here to follow the link

Rights:

Biblioteka Uniwersytetu Zielonogórskiego

Subject and keywords:

skin lesion feature extraction graph theory multi-class prediction SHAP values

Object collections:

Last modified:

Aug 5, 2025

In our library since:

Aug 5, 2025

Number of object content hits:

12

All available object's versions:

https://zbc.uz.zgora.pl/repozytorium/publication/101891

Show description in RDF format:

RDF

Show description in OAI-PMH format:

OAI-PMH

Edition name Date
Enhancing multi-class prediction of skin lesions with feature importance assessment Aug 5, 2025

Objects Similar

AMCS, Volume 27 (2017)
AMCS, Volume 27 (2017)
On the predictive power of meta-features in OpenML

Bilalli, Besim Abelló, Alberto Aluja-Banet, Tomas Stefanowski, Jerzy - ed. Krawiec, Krzysztof - ed. Wrembel, Robert - ed.

2017
artykuł
AMCS, Volume 15 (2005)
AMCS, Volume 15 (2005)
Image retrieval based on hierarchical gabor filters

Andrysiak, Tomasz Choraś, Michał Beliczyński, Bartłomiej - red.

2005
artykuł
AMCS, Volume 24 (2014)
AMCS, Volume 24 (2014)
A simultaneous localization and tracking method for a worm tracking system

Kowalski, Mateusz Kaczmarek, Piotr Kabaciński, Rafał Matuszczak, Mieszko Tranbowicz, Kamil Sobkowiak, Robert Abaev, Pavel - ed. Razumchik, Rostislav - ed. Kołodziej, Joanna - ed.

2014
artykuł
AMCS, volume 34 (2024)
AMCS, volume 34 (2024)
A review of shockable arrhythmia detection of ECG signals using machine and deep learning techniques

Kavya, Lakkakula Karuna, Yepuganti Saritha, Saladi Prakash, Allam Jaya Patro, Kiran Kumar Sahoo, Suraj Prakash Tadeusiewicz, Ryszard (1947- ) Pławiak, Paweł Korbicz, Józef (1951- ) - red. Uciński, Dariusz - red.

2024
artykuł
AMCS, volume 28 (2018)
AMCS, volume 28 (2018)
Clustering based on eigenvectors of the adjacency matrix

Lucińska, Małgorzata Wierzchoń, Sławomir T. Korbicz, Józef (1951- ) - red. Uciński, Dariusz - red.

2018
artykuł
More
×

Citation

Citation style:

Contact details

Address

University Library
al. Wojska Polskiego 71
65-762 Zielona Góra

Phone

(+48) 68 328 21 55

E-Mail

repozytorium@uz.zgora.pl

Visit us!

This page uses 'cookies'. More information