Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification

Abstract Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths worldwide, necessitating accurate and efficient diagnostic tools. This study proposes an advanced deep learning-based framework for classifying CRC histological images into eight categories: Adipose, Complex,...

Full description

Saved in:
Bibliographic Details
Main Authors: Pragati Patharia, Prabira Kumar Sethy, K. Lakshmipathi Raju, Anita Khanna, Ashoka Kumar Ratha, Santi Kumari Behera, Aziz Nanthaamornphong
Format: Article
Language:English
Published: Springer 2025-07-01
Series:Discover Artificial Intelligence
Subjects:
Online Access:https://doi.org/10.1007/s44163-025-00404-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849235150587559936
author Pragati Patharia
Prabira Kumar Sethy
K. Lakshmipathi Raju
Anita Khanna
Ashoka Kumar Ratha
Santi Kumari Behera
Aziz Nanthaamornphong
author_facet Pragati Patharia
Prabira Kumar Sethy
K. Lakshmipathi Raju
Anita Khanna
Ashoka Kumar Ratha
Santi Kumari Behera
Aziz Nanthaamornphong
author_sort Pragati Patharia
collection DOAJ
description Abstract Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths worldwide, necessitating accurate and efficient diagnostic tools. This study proposes an advanced deep learning-based framework for classifying CRC histological images into eight categories: Adipose, Complex, Debris, Empty, Lympho, Mucosa, Stroma, and Tumor. Initially, 18 pre-trained Convolutional Neural Networks (CNNs) were evaluated using transfer learning techniques. Statistical analysis via Duncan's Multirange test identified Darknet53 as the most effective model, achieving an accuracy (Acc.) of 92.43%, sensitivity (Sen.) of 92.43%, specificity (Spec.) of 98.51%, precision (Prec.) of 92.70%, and F1-score of 92.45%. To enhance the classification performance, Darknet53 was hybridized with a SVM by replacing the dense layer, and hyperparameters were optimized using a Random Grid Search algorithm. The optimized hybrid model exhibited a remarkable improvement, with an Acc. of 99.7%, Sen. of 99.7%, Spec. of 99.91%, Prec. of 99.98%, and F1-score of 99.98%, alongside significant improvements in other metrics. This hybrid approach demonstrates substantial potential in automating CRC diagnosis with high Acc., offering a valuable tool for clinical use.
format Article
id doaj-art-98f7db8208fb4d0ca919c650ad1de7e6
institution Kabale University
issn 2731-0809
language English
publishDate 2025-07-01
publisher Springer
record_format Article
series Discover Artificial Intelligence
spelling doaj-art-98f7db8208fb4d0ca919c650ad1de7e62025-08-20T04:02:54ZengSpringerDiscover Artificial Intelligence2731-08092025-07-015111510.1007/s44163-025-00404-8Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classificationPragati Patharia0Prabira Kumar Sethy1K. Lakshmipathi Raju2Anita Khanna3Ashoka Kumar Ratha4Santi Kumari Behera5Aziz Nanthaamornphong6Department of ECE, Guru Ghasidas VishwavidyalayaDepartment of ElectronicsDepartment of IT, S.R.K.R Engineering CollegeDepartment of ECE, Guru Ghasidas VishwavidyalayaDepartment of ElectronicsDepartment of Computer Science and Engineering, VSSUTCollege of Computing, Prince of Songkla University, Phuket CampusAbstract Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths worldwide, necessitating accurate and efficient diagnostic tools. This study proposes an advanced deep learning-based framework for classifying CRC histological images into eight categories: Adipose, Complex, Debris, Empty, Lympho, Mucosa, Stroma, and Tumor. Initially, 18 pre-trained Convolutional Neural Networks (CNNs) were evaluated using transfer learning techniques. Statistical analysis via Duncan's Multirange test identified Darknet53 as the most effective model, achieving an accuracy (Acc.) of 92.43%, sensitivity (Sen.) of 92.43%, specificity (Spec.) of 98.51%, precision (Prec.) of 92.70%, and F1-score of 92.45%. To enhance the classification performance, Darknet53 was hybridized with a SVM by replacing the dense layer, and hyperparameters were optimized using a Random Grid Search algorithm. The optimized hybrid model exhibited a remarkable improvement, with an Acc. of 99.7%, Sen. of 99.7%, Spec. of 99.91%, Prec. of 99.98%, and F1-score of 99.98%, alongside significant improvements in other metrics. This hybrid approach demonstrates substantial potential in automating CRC diagnosis with high Acc., offering a valuable tool for clinical use.https://doi.org/10.1007/s44163-025-00404-8Colorectal cancer classificationDarknet53SVMRandom grid searchHistological image analysis
spellingShingle Pragati Patharia
Prabira Kumar Sethy
K. Lakshmipathi Raju
Anita Khanna
Ashoka Kumar Ratha
Santi Kumari Behera
Aziz Nanthaamornphong
Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
Discover Artificial Intelligence
Colorectal cancer classification
Darknet53
SVM
Random grid search
Histological image analysis
title Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
title_full Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
title_fullStr Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
title_full_unstemmed Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
title_short Hybrid Darknet53-SVM model with random grid search optimization for enhanced colorectal cancer histological image classification
title_sort hybrid darknet53 svm model with random grid search optimization for enhanced colorectal cancer histological image classification
topic Colorectal cancer classification
Darknet53
SVM
Random grid search
Histological image analysis
url https://doi.org/10.1007/s44163-025-00404-8
work_keys_str_mv AT pragatipatharia hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT prabirakumarsethy hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT klakshmipathiraju hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT anitakhanna hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT ashokakumarratha hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT santikumaribehera hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification
AT aziznanthaamornphong hybriddarknet53svmmodelwithrandomgridsearchoptimizationforenhancedcolorectalcancerhistologicalimageclassification