Melanoma Skin Cancer Detection with the Integration of a Conversational Chatbot

Abstract

Skin cancer, specifically melanoma, results from abnormal melanocytic cell growth and can be fatal. It typically appears as dark lesions due to UV exposure and genetic factors. Early detection is crucial for treatment. The conventional method, biopsy, is invasive, painful, and slow, as it requires lab analysis. To address these issues, a non-invasive computer-aided diagnosis (CAD) system is proposed, using dermoscopy images. This system preprocesses the images, segments the lesion, extracts unique features, and then classifies the skin as normal or cancerous using a support vector machine (SVM). The SVM with a linear kernel demonstrates optimal accuracy. CAD eliminates the need for physical contact, reducing pain and improving efficiency in melanoma detection through advanced image processing techniques

Introduction

Overview

Cancer, especially melanoma (a type of skin cancer), is one of the leading causes of death globally. It arises from the uncontrolled growth of abnormal skin cells, often triggered by DNA defects. Melanoma is particularly dangerous, often mimicking benign moles in early stages but distinguished by asymmetry, irregular borders, color variation, and larger diameter (>6mm).

Prevalence and Impact

• Skin cancer is highly prevalent in fair-skinned individuals and increasingly affects young people, especially women.

• In 2022, ~9,730 melanoma-related deaths were estimated.

• Early diagnosis is crucial, prompting the development of Computer-Assisted Diagnostic (CAD) tools using digital dermoscopic images.

Technology in Detection

To improve accuracy and speed, AI-based systems are being developed that use:

• Image processing (e.g., segmentation, feature extraction)

• Machine learning classifiers like CNN (Convolutional Neural Networks) and SVM (Support Vector Machines)

• Chatbot integration for patient assistance

Image Analysis Pipeline:

1. Image Acquisition: Collect skin lesion images.

2. Preprocessing: Remove noise, blur, and irrelevant details.

3. Feature Extraction: Analyze shape, symmetry, color, and texture.

4. Segmentation: Isolate the lesion area.

5. Classification: Use ML algorithms (CNN, SVM) to detect melanoma.

6. Result Interpretation: Determine if the lesion is cancerous or benign.

Literature Review Insights

Various studies have implemented:

• Color correlograms, Otsu thresholding, active contour models

• Bayesian, SVM, and Decision Tree classifiers

• Features used: Shape, color, geometry, texture

• Outcomes show non-invasive, automated systems outperform manual diagnosis in terms of speed and reliability.

System Design

• User Interaction Layer: Register/login, upload images, receive results

• Database: Labeled image data for training/testing

• Chatbot: Assists users in understanding skin cancer and using the system

• Real-time analysis for instant results and guidance

Model Features and Benefits

1. High Accuracy: Combining CNN for pattern learning with SVM for fine classification.

2. Early Detection: Enables prompt treatment and reduces mortality.

3. User-Friendly: Simple interface with chatbot support.

4. Self-Automated: Reduces reliance on medical personnel for screening.

5. Cost-Effective: Reduces need for invasive diagnostics like biopsies.

6. Adaptable: Model improves over time with more data.

7. Fast Results: Provides immediate feedback to users and doctors.

Conclusion

In conclusion, the two-stage model combining CNN and SVM algorithms proves to be an effective approach for classifying skin lesions into melanoma and non-melanoma types. The CNN excels in feature extraction and initial classification by learning deep image characteristics, while the SVM refines these results, leading to enhanced prediction accuracy. This hybrid model combines the advantages of CNN and SVM which ensures fast, accurate and dependable melanoma detection. The use of a diverse dataset to train the system ensures good performance, and therefore the model will help in early detection of melanoma which can result in timely and life saving medical treatment.

References

[1] L. Yu, H. Chen, Q. Dou, J. Qin and P. Heng, \"Automated Melanoma Recognition in Dermoscopy Images via Very Deep Residual Networks,\" in IEEE Transactions on Medical Imaging, April 2017. [2] Naeem, Ahmad, Shoaib Farooq, Adel Khelifi and Adnan Abid. “Malignant Melanoma Classification Using Deep Learning: Datasets, Performance Measurements, Challenges and Opportunities T.” Department of Computer Science, University of Management and Technology Lahore, Pakistan; Abu Dhabi University, United Arab Emirates, IEEE, 2017 [3] Farooq MA, Raza RH and Azhar MAM \"Automatic Lesion Detection System (ALDS) for Skin Cancer Classification Using SVM and Neural Networks.\" 16th IEEE International Conference on Bioinformatics and Bioengineering, Proceedings, IEEE, 2016. [4] Sreedhar B, Swamy MB and Kumar MS “Comparative Analysis of Melanoma skin Cancer Detection using conventional and modern image processing methods” 11 Feb 2018. [5] Gupta, A., Thakur, S., & Rana, A. \"A Study on Techniques for Melanoma Detection and Classification.\" Amity School of Engineering and Technology, Amity University Uttar Pradesh, 2020 [6] Kavitha, P., & Jayalakshmi, V. \"A Survey on Skin Cancer Detection Using Various Image Processing Techniques.\" Proceedings of the Third International Conference on Intelligent Sustainable Systems (ICISS), IEEE Xplore, 2020 [7] Selvarasa, M., & Aponso, A. \"Critical Review of Computer-Aided Techniques for Skin Cancer Screening.\" Proceedings of the International Conference on Engineering and IT, IEEE Xplore, 2020 [8] Ahmed Thaajwer M.A., UA. Piumi Ishanka \"Melanoma Skin Cancer Detection Through Image Processing and Machine Learning Techniques.\" Department of Computing & Information Systems, Faculty of Applied Sciences, Sabaragamuwa University of Sri Lanka, presented at the 2020 [9] Kaur, R., GholamHosseini, H., Sinha, R., & Lindén,M. (2022). Melanoma Classification Using a Novel Deep Convolutional Neural Network with Dermoscopic Images.

Copyright

Copyright © 2025 Bhagyashree Kadam, Ritesh Kolhe, Sagar Gonjare , Rameshwar Shinde, Sanket Yelam. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.