DEEP LEARNING FOR HAZARDOUS CHEMICALS INDUCED SKIN DISEASES CLASSIFICATION: A COMPREHENSIVE ANALYSIS ON INCEPTIONV3 ARCHITECTURE AND MODEL EVALUATION
Abstract
This research investigates the utilization of deep learning with focus on the InceptionV3 architecture for the categorization of skin conditions induced by contact with hazardous chemicals. The model trained using a carefully curated dataset comprising images of various skin diseases achieves an overall prediction accuracy of 66.67% successfully identifying four out of six evaluated classes. While promising potential practical applications, the analysis reveals a tendency towards over fitting, evidenced by the discrepancy between training and validation accuracies and losses. Future research should focus on mitigating over fitting and expanding the dataset to enhance the model’s generalizability and reliability across a diverse range of cases.

Authors
Amar T. Banmare1, Nitin K. Choudhari2, Archana R. Chaudhari3
Rashtrasant Tukadoji Maharaj Nagpur University, India1, Priyadarshini Bhagwati College of Engineering, India 2,3

Keywords
Skin Disease Classification, Deep Learning, InceptionV3, Hazardous Chemical, Model Evaluation
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
000300000000
Published By :
ICTACT
Published In :
ICTACT Journal on Data Science and Machine Learning
( Volume: 5 , Issue: 2 , Pages: 559 - 565 )
Date of Publication :
March 2024
Page Views :
33
Full Text Views :
3

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.