Irfan Sadiq Rahat
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Irfan Sadiq Rahat has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
Hritwik Ghosh
Vellore Institute of Technology University
5 shared publications- Hritwik Ghosh
Vellore Institute of Technology University
3 shared publications Syed Khasim
Vellore Institute of Technology University
3 shared publicationsKareemulla Shaik
Vellore Institute of Technology University
2 shared publicationsManava Yesubabu
Vardhaman College of Engineering
2 shared publications- Sachi Nandan Mohanty
School of Computer Science and Engineering (SCOPE), VIT-AP University, Amaravati, India
2 shared publications
Latest Publications
- Open Access
- 10 November 2026
- Published by BON VIEW PUBLISHING PTE in Medinformatics
Automated Pharyngitis Classification Using a Dual-Branch Hybrid Deep Learning Model with Explainable Artificial Intelligence and Retrieval-Augmented Generation-Based Clinical Recommendation
Pharyngitis is a frequent inflammatory condition of the upper-airway system for which prompt preliminary evaluation can aid in the proper clinical evaluation and minimize treatment outside of that. However, the level of inflammation, the quality of illumination, the quality of the image, and the level of the observer’s experience can affect the interpretation of the image. A research-stage framework for binary pharyngitis screening with 312 pharyngitis images classified as no pharyngitis and pharyngitis is presented. The proposed dual-branch hybrid deep learning model consists of a separable convolution-based local feature extraction branch and an attention-based global contextual learning and cross-scale feature fusion branch. It brings together the combined power of hybrid CNN/attention based classification, explainable AI, and retrieval-augmented generation, along with a Qwen large language model, in a single screening workflow. The model performance was done by 5-fold cross-validation with leakage-safe and the result is 97.25% accuracy, 97.58% precision, 97.43% recall, and 97.50% F1 score. To explain the predictions, image regions were emphasized by explainability visualization, and the RAG/Qwen module produced structured recommendation reports using the predicted class and retrieved medical context. The framework can help with initial image screening, and clinician-reviewed decision support for telemedicine/triage environments. However, the limited sample size and absence of external multi-center and prospective testing by clinicians with the weathered data limits clinical generalizability. Thus, the proposed system is not intended as a diagnostic or a prescription tool for treatment, but as a decision support system in a research phase. Received: 24 April 2026 | Revised: 27 July 2026 | Accepted: 15 September 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement We confirm that the image datasets described in this study are publicly available in the following repositories: 1. Pharyngitis Images Dataset Kaggle: https://www.kaggle.com/datasets/bebeaa/pharyngitis-images-dataset. 2. PGUPharyngitis Figshare: https://doi.org/10.6084/m9.figshare.28163513. Author Contribution Statement Mizanur Rahman: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft. Fakrul Alam: Validation, Formal analysis, Data curation. Azharul Islam: Methodology, Validation, Formal analysis. Mazharul Islam: Data curation, Writing – review & editing. Sumon Rana: Formal analysis, Writing – review & editing. Firoz Kabir: Validation, Formal analysis, Writing – review & editing. Irfan Sadiq Rahat: Software, Writing – original draft, Supervision.
- 10 November 2026
- Published by BON VIEW PUBLISHING PTE in Medinformatics
Open AccessAutomated Pharyngitis Classification Using a Dual-Branch Hybrid Deep Learning Model with Explainable Artificial Intelligence and Retrieval-Augmented Generation-Based Clinical Recommendation
Pharyngitis is a frequent inflammatory condition of the upper-airway system for which prompt preliminary evaluation can aid in the proper clinical evaluation and minimize treatment outside of that. However, the level of inflammation, the quality of illumination, the quality of the image, and the level of the observer’s experience can affect the interpretation of the image. A research-stage framework for binary pharyngitis screening with 312 pharyngitis images classified as no pharyngitis and pharyngitis is presented. The proposed dual-branch hybrid deep learning model consists of a separable convolution-based local feature extraction branch and an attention-based global contextual learning and cross-scale feature fusion branch. It brings together the combined power of hybrid CNN/attention based classification, explainable AI, and retrieval-augmented generation, along with a Qwen large language model, in a single screening workflow. The model performance was done by 5-fold cross-validation with leakage-safe and the result is 97.25% accuracy, 97.58% precision, 97.43% recall, and 97.50% F1 score. To explain the predictions, image regions were emphasized by explainability visualization, and the RAG/Qwen module produced structured recommendation reports using the predicted class and retrieved medical context. The framework can help with initial image screening, and clinician-reviewed decision support for telemedicine/triage environments. However, the limited sample size and absence of external multi-center and prospective testing by clinicians with the weathered data limits clinical generalizability. Thus, the proposed system is not intended as a diagnostic or a prescription tool for treatment, but as a decision support system in a research phase. Received: 24 April 2026 | Revised: 27 July 2026 | Accepted: 15 September 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement We confirm that the image datasets described in this study are publicly available in the following repositories: 1. Pharyngitis Images Dataset Kaggle: https://www.kaggle.com/datasets/bebeaa/pharyngitis-images-dataset. 2. PGUPharyngitis Figshare: https://doi.org/10.6084/m9.figshare.28163513. Author Contribution Statement Mizanur Rahman: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing – original draft. Fakrul Alam: Validation, Formal analysis, Data curation. Azharul Islam: Methodology, Validation, Formal analysis. Mazharul Islam: Data curation, Writing – review & editing. Sumon Rana: Formal analysis, Writing – review & editing. Firoz Kabir: Validation, Formal analysis, Writing – review & editing. Irfan Sadiq Rahat: Software, Writing – original draft, Supervision.
- 13 August 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 180, 108962
Advancing breast ultrasound diagnostics through hybrid deep learning models
- 13 August 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 180, 108962
Advancing breast ultrasound diagnostics through hybrid deep learning models
- Open Access
- 26 March 2024
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 10
Convolutional Neural Networks in Malaria Diagnosis: A Study on Cell Image Classification
INTRODUCTION: Malaria, a persistent global health threat caused by Plasmodium parasites, necessitates rapid and accurate identification for effective treatment and containment. This study investigates the utilization of convolutional neural networks (CNNs) to enhance the precision and speed of malaria detection through the classification of cell images infected with malaria.OBJECTIVES: The primary objective of this research is to explore the effectiveness of CNNs in accurately classifying malaria-infected cell images. By employing various deep learning models, including ResNet50, AlexNet, Inception V3, VGG19, VGG16, and MobileNetV2, the study aims to assess the performance of each model and identify their strengths and weaknesses in malaria diagnosis.METHODS: A balanced dataset comprising approximately 8,000 enhanced images of blood cells, evenly distributed between infected and uninfected classes, was utilized for model training and evaluation. Performance evaluation metrics such as precision, recall, F1-score, and accuracy were employed to assess the efficacy of each CNN model in malaria classification.RESULTS: The results demonstrate high accuracy across all models, with AlexNet and VGG19 exhibiting the highest levels of accuracy. However, the selection of a model should consider specific application requirements and constraints, as each model presents unique trade-offs between computational efficiency and performance.CONCLUSION: This study contributes to the burgeoning field of deep learning in healthcare, particularly in utilizing medical imaging for disease diagnosis. The findings underscore the considerable potential of CNNs in enhancing malaria diagnosis. Future research directions may involve further model optimization, exploration of larger and more diverse datasets, and the integration of CNNs into practical diagnostic tools for real-world deployment.
- 26 March 2024
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 10
Open AccessConvolutional Neural Networks in Malaria Diagnosis: A Study on Cell Image Classification
INTRODUCTION: Malaria, a persistent global health threat caused by Plasmodium parasites, necessitates rapid and accurate identification for effective treatment and containment. This study investigates the utilization of convolutional neural networks (CNNs) to enhance the precision and speed of malaria detection through the classification of cell images infected with malaria.OBJECTIVES: The primary objective of this research is to explore the effectiveness of CNNs in accurately classifying malaria-infected cell images. By employing various deep learning models, including ResNet50, AlexNet, Inception V3, VGG19, VGG16, and MobileNetV2, the study aims to assess the performance of each model and identify their strengths and weaknesses in malaria diagnosis.METHODS: A balanced dataset comprising approximately 8,000 enhanced images of blood cells, evenly distributed between infected and uninfected classes, was utilized for model training and evaluation. Performance evaluation metrics such as precision, recall, F1-score, and accuracy were employed to assess the efficacy of each CNN model in malaria classification.RESULTS: The results demonstrate high accuracy across all models, with AlexNet and VGG19 exhibiting the highest levels of accuracy. However, the selection of a model should consider specific application requirements and constraints, as each model presents unique trade-offs between computational efficiency and performance.CONCLUSION: This study contributes to the burgeoning field of deep learning in healthcare, particularly in utilizing medical imaging for disease diagnosis. The findings underscore the considerable potential of CNNs in enhancing malaria diagnosis. Future research directions may involve further model optimization, exploration of larger and more diverse datasets, and the integration of CNNs into practical diagnostic tools for real-world deployment.
- 7 March 2024
- Published by Springer Nature
- p. 97-110
Benchmarking ML and DL Models for Mango Leaf Disease Detection: A Comparative Analysis
- 7 March 2024
- Published by Springer Nature
- p. 97-110
Benchmarking ML and DL Models for Mango Leaf Disease Detection: A Comparative Analysis
- Open Access
- 12 January 2024
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Internet of Things
- Vol. 10
Enhancing Agricultural Sustainability with Deep Learning: A Case Study of Cauliflower Disease Classification
The pivotal role of sustainable agriculture in ensuring food security and nurturing healthy farming communities is undeniable. Among the numerous challenges encountered in this domain, one key hurdle is the early detection and effective treatment of diseases impacting crops, specifically cauliflower.This research provides an in-depth exploration of the use of advanced DL algorithms to perform efficient identification and classification of cauliflower diseases. The study employed and scrutinized four leading DL models: EfficientNetB3, DenseNet121, VGG19 CNN, and ResNet50, assessing their capabilities based on the accuracy of disease detection.The investigation revealed a standout performer, the EfficientNetB3 model, which demonstrated an exceptional accuracy rate of 98%. The remaining models also displayed commendable performance, with DenseNet121 and VGG19 CNN attaining accuracy rates of 81% and 84%, respectively, while ResNet50 trailed at 78%. The noteworthy performance of the EfficientNetB3 model is indicative of its vast potential to contribute to agricultural sustainability. Its ability to detect and classify cauliflower diseases accurately and promptly allows for early interventions, reducing the risk of extensive crop damage.This study contributes valuable insights to the expanding field of DL applications in agriculture. These findings are expected to guide the development of advanced agricultural monitoring systems and decision-support tools, ultimately fostering a more sustainable and productive agricultural landscape.
- 12 January 2024
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Internet of Things
- Vol. 10
Open AccessEnhancing Agricultural Sustainability with Deep Learning: A Case Study of Cauliflower Disease Classification
The pivotal role of sustainable agriculture in ensuring food security and nurturing healthy farming communities is undeniable. Among the numerous challenges encountered in this domain, one key hurdle is the early detection and effective treatment of diseases impacting crops, specifically cauliflower.This research provides an in-depth exploration of the use of advanced DL algorithms to perform efficient identification and classification of cauliflower diseases. The study employed and scrutinized four leading DL models: EfficientNetB3, DenseNet121, VGG19 CNN, and ResNet50, assessing their capabilities based on the accuracy of disease detection.The investigation revealed a standout performer, the EfficientNetB3 model, which demonstrated an exceptional accuracy rate of 98%. The remaining models also displayed commendable performance, with DenseNet121 and VGG19 CNN attaining accuracy rates of 81% and 84%, respectively, while ResNet50 trailed at 78%. The noteworthy performance of the EfficientNetB3 model is indicative of its vast potential to contribute to agricultural sustainability. Its ability to detect and classify cauliflower diseases accurately and promptly allows for early interventions, reducing the risk of extensive crop damage.This study contributes valuable insights to the expanding field of DL applications in agriculture. These findings are expected to guide the development of advanced agricultural monitoring systems and decision-support tools, ultimately fostering a more sustainable and productive agricultural landscape.
- Open Access
- 19 December 2023
- Published by MDPI AG in Engineering Proceedings
- Vol. 59 (1) , 69
Advanced Deep Learning Models for Corn Leaf Disease Classification: A Field Study in Bangladesh
Agriculture is pivotal in Bangladesh, with maize being a central crop. However, leaf diseases significantly threaten its productivity. This study introduces deep learning models for enhanced disease detection in maize. We developed an unique datasets of 4800 maize leaf images, categorized into four health conditions: Healthy, Common Rust, Gray Leaf Spot, and Blight. These images underwent extensive Pre-processing and data augmentation to improve robustness. We explored various deep learning models, including ResNet50GAP, DenseNet121, VGG19, and a custom Sequential model. DenseNet121 and VGG19 showed exceptional performance, achieving accuracies of 99.22% and 99.44% respectively. Our research is novel due to the integration of transfer learning and image augmentation, enhancing the models’ generalization capabilities. A hybrid model combining ResNet50 and VGG16 features achieved a remarkable 99.65% accuracy, validating our approach. Our results indicate that deep learning can significantly impact agricultural diagnostics, offering new research directions and applications. This study highlights the potential artificial intelligence in advancing agricultural practices and food security in Bangladesh, emphasizing the need for model interpretability to build trust in machine learning solutions.
- 19 December 2023
- Published by MDPI AG in Engineering Proceedings
- Vol. 59 (1) , 69
Open AccessAdvanced Deep Learning Models for Corn Leaf Disease Classification: A Field Study in Bangladesh
Agriculture is pivotal in Bangladesh, with maize being a central crop. However, leaf diseases significantly threaten its productivity. This study introduces deep learning models for enhanced disease detection in maize. We developed an unique datasets of 4800 maize leaf images, categorized into four health conditions: Healthy, Common Rust, Gray Leaf Spot, and Blight. These images underwent extensive Pre-processing and data augmentation to improve robustness. We explored various deep learning models, including ResNet50GAP, DenseNet121, VGG19, and a custom Sequential model. DenseNet121 and VGG19 showed exceptional performance, achieving accuracies of 99.22% and 99.44% respectively. Our research is novel due to the integration of transfer learning and image augmentation, enhancing the models’ generalization capabilities. A hybrid model combining ResNet50 and VGG16 features achieved a remarkable 99.65% accuracy, validating our approach. Our results indicate that deep learning can significantly impact agricultural diagnostics, offering new research directions and applications. This study highlights the potential artificial intelligence in advancing agricultural practices and food security in Bangladesh, emphasizing the need for model interpretability to build trust in machine learning solutions.
- Open Access
- 27 November 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Internet of Things
- Vol. 10
Deciphering Microorganisms through Intelligent Image Recognition: Machine Learning and Deep Learning Approaches, Challenges, and Advancements
Microorganisms are pervasive and have a significant impact in various fields such as healthcare, environmental monitoring, and biotechnology. Accurate classification and identification of microorganisms are crucial for professionals in diverse areas, including clinical microbiology, agriculture, and food production. Traditional methods for analyzing microorganisms, like culture techniques and manual microscopy, can be labor-intensive, expensive, and occasionally inadequate due to morphological similarities between different species. As a result, there is an increasing need for intelligent image recognition systems to automate microorganism classification procedures with minimal human involvement. In this paper, we present an in-depth analysis of ML and DL perspectives used for the precise recognition and classification of microorganism images, utilizing a dataset comprising eight distinct microorganism types: Spherical bacteria, Amoeba, Hydra, Paramecium, Rod bacteria, Spiral bacteria, Euglena and Yeast. We employed several ml algorithms including SVM, Random Forest, and KNN, as well as the deep learning algorithm CNN. Among these methods, the highest accuracy was achieved using the CNN approach. We delve into current techniques, challenges, and advancements, highlighting opportunities for further progress.
- 27 November 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Internet of Things
- Vol. 10
Open AccessDeciphering Microorganisms through Intelligent Image Recognition: Machine Learning and Deep Learning Approaches, Challenges, and Advancements
Microorganisms are pervasive and have a significant impact in various fields such as healthcare, environmental monitoring, and biotechnology. Accurate classification and identification of microorganisms are crucial for professionals in diverse areas, including clinical microbiology, agriculture, and food production. Traditional methods for analyzing microorganisms, like culture techniques and manual microscopy, can be labor-intensive, expensive, and occasionally inadequate due to morphological similarities between different species. As a result, there is an increasing need for intelligent image recognition systems to automate microorganism classification procedures with minimal human involvement. In this paper, we present an in-depth analysis of ML and DL perspectives used for the precise recognition and classification of microorganism images, utilizing a dataset comprising eight distinct microorganism types: Spherical bacteria, Amoeba, Hydra, Paramecium, Rod bacteria, Spiral bacteria, Euglena and Yeast. We employed several ml algorithms including SVM, Random Forest, and KNN, as well as the deep learning algorithm CNN. Among these methods, the highest accuracy was achieved using the CNN approach. We delve into current techniques, challenges, and advancements, highlighting opportunities for further progress.
- Open Access
- 3 October 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 9
An All-Inclusive Machine Learning and Deep Learning Method for Forecasting Cardiovascular Disease in Bangladeshi Population
INTRODUCTION: Cardiovascular disease is a major concern and pressing issue faced by the healthcare sector globally. According to a survey conducted by the WHO every year, CVDs cause 17.9 million deaths worldwide. Lack of pre-prediction of CVDs is a significant factor contributing to the death of patients. Predicting CVDs is a challenging task for medical practitioners as it requires a high level of medical analysis skills and extensive knowledge. OBJECTIVES: We believe that the improvement in the accuracy of prediction can significantly reduce the risk caused by CVDs and help medical practitioners better diagnose patients . METHODS: In this study, We created a CVD prediction model. using a ML approach. We utilized various algorithms, including logistic regression, Gaussian Naive Baye, Bernoulli Naive Baye, SVM, KNN, optimized KNN, X Gradient Boosting, and random forest algorithms to analyze and predict CVDs. RESULTS: Our developed prediction model achieved an accuracy of 96.7%, indicating its effectiveness in predicting CVDs. DL algorithms can also assist in identifying, classifying, and quantifying patterns of medical images, improving patient evaluation and diagnosis based on prior medical history and evaluation patterns. CONCLUSION: Furthermore, deep learning algorithms can help in developing new drugs with minimum cost by reducing the number of clinical research trials, using prior prediction of the drug's efficacy.
- 3 October 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 9
Open AccessAn All-Inclusive Machine Learning and Deep Learning Method for Forecasting Cardiovascular Disease in Bangladeshi Population
INTRODUCTION: Cardiovascular disease is a major concern and pressing issue faced by the healthcare sector globally. According to a survey conducted by the WHO every year, CVDs cause 17.9 million deaths worldwide. Lack of pre-prediction of CVDs is a significant factor contributing to the death of patients. Predicting CVDs is a challenging task for medical practitioners as it requires a high level of medical analysis skills and extensive knowledge. OBJECTIVES: We believe that the improvement in the accuracy of prediction can significantly reduce the risk caused by CVDs and help medical practitioners better diagnose patients . METHODS: In this study, We created a CVD prediction model. using a ML approach. We utilized various algorithms, including logistic regression, Gaussian Naive Baye, Bernoulli Naive Baye, SVM, KNN, optimized KNN, X Gradient Boosting, and random forest algorithms to analyze and predict CVDs. RESULTS: Our developed prediction model achieved an accuracy of 96.7%, indicating its effectiveness in predicting CVDs. DL algorithms can also assist in identifying, classifying, and quantifying patterns of medical images, improving patient evaluation and diagnosis based on prior medical history and evaluation patterns. CONCLUSION: Furthermore, deep learning algorithms can help in developing new drugs with minimum cost by reducing the number of clinical research trials, using prior prediction of the drug's efficacy.
- Open Access
- 29 September 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 9
Unraveling the Heterogeneity of Lower-Grade Gliomas: Deep Learning-Assisted Flair Segmentation and Genomic Analysis of Brain MR Images
The precise identification of FLAIR abnormalities in brain MR images is essential for diagnosing and managing lower-grade gliomas, segmentation continues to be a difficult task. In this research, we develop an exhaustive strategy that integrates advanced deep learning models such as DeepLabv3, U-Net, DenseNet121-Unet, ResNet50, Attention U-Net and EfficientNet to effectively segment FLAIR abnormalities in a dataset comprising 110 lower-grade glioma patients. The cancer Imaging achieve (TCIA), includes genomic cluster data and patient-specific details. Our methodology tackles the multi-class data imbalanced by employing a customized loss function, which merges Categorical Cross Entropy (CCE) WCE and WMDL functions are used to calculate loss, allowing the network to accurately segment smaller tumor regions. By performing dense network training on 3D picture patches, the suggested technique improves detection of border region artifacts and efficiently manages storage and system limited resources. We evaluate our strategy’s effectiveness on the presented dataset, emphasizing its potential for assisting correct diagnosis and individualized treatment strategies for patients with lower-grade gliomas.
- 29 September 2023
- Published by European Alliance for Innovation n.o. in EAI Endorsed Transactions on Pervasive Health and Technology
- Vol. 9
Open AccessUnraveling the Heterogeneity of Lower-Grade Gliomas: Deep Learning-Assisted Flair Segmentation and Genomic Analysis of Brain MR Images
The precise identification of FLAIR abnormalities in brain MR images is essential for diagnosing and managing lower-grade gliomas, segmentation continues to be a difficult task. In this research, we develop an exhaustive strategy that integrates advanced deep learning models such as DeepLabv3, U-Net, DenseNet121-Unet, ResNet50, Attention U-Net and EfficientNet to effectively segment FLAIR abnormalities in a dataset comprising 110 lower-grade glioma patients. The cancer Imaging achieve (TCIA), includes genomic cluster data and patient-specific details. Our methodology tackles the multi-class data imbalanced by employing a customized loss function, which merges Categorical Cross Entropy (CCE) WCE and WMDL functions are used to calculate loss, allowing the network to accurately segment smaller tumor regions. By performing dense network training on 3D picture patches, the suggested technique improves detection of border region artifacts and efficiently manages storage and system limited resources. We evaluate our strategy’s effectiveness on the presented dataset, emphasizing its potential for assisting correct diagnosis and individualized treatment strategies for patients with lower-grade gliomas.
- Open Access
- 21 September 2023
- Published by European Alliance for Innovation n.o. in ICST Transactions on Scalable Information Systems
- Vol. 10 (6)
Potato Leaf Disease Recognition and Prediction using Convolutional Neural Networks
Potato crops are vital to global food security and economy, yet they are vulnerable to a wide range of leaf diseases that can significantly impact yield and quality. Rapid diagnosis and accurate identification of these disorders are critical for effective disease control and prevention. In this research, we offer an extensive evaluation and contrast of three state -of-art CNN models- VGG19, DenseNet121 and ResNet50-in order to identify and forecast potato leaf diseases. Our study employed a sizable dataset of potato leaf images, containing diverse healthy and afflicted specimens, to train and assess the performance of the chosen CNN models. Extensive data augmentation techniques were employed to enhance the dataset’s diversity and generalization capabilities. We evaluated the models considering their accuracy, precision, recall, F1-score and computational efficiency to determine the most fitting model for real-life applications. The results demonstrate that all three CNN models achieved high performance in identifying and predicting potato leaf diseases, with VGG19 emerging as the top performer followed closely by DenseNet121 and ResNet50.Our findings provide valuable insights into the efficacy of DL approaches for potato leaf ailment identification and offer a foundation for future research and deployment of these models in precision agriculture systems. Ultimately, this work aims to support the development of more robust and efficient tools for timely disease diagnosis, enabling farmers and agronomists to make better-informed decisions and safeguard the health and productivity of potato crops worldwide.
- 21 September 2023
- Published by European Alliance for Innovation n.o. in ICST Transactions on Scalable Information Systems
- Vol. 10 (6)
Open AccessPotato Leaf Disease Recognition and Prediction using Convolutional Neural Networks
Potato crops are vital to global food security and economy, yet they are vulnerable to a wide range of leaf diseases that can significantly impact yield and quality. Rapid diagnosis and accurate identification of these disorders are critical for effective disease control and prevention. In this research, we offer an extensive evaluation and contrast of three state -of-art CNN models- VGG19, DenseNet121 and ResNet50-in order to identify and forecast potato leaf diseases. Our study employed a sizable dataset of potato leaf images, containing diverse healthy and afflicted specimens, to train and assess the performance of the chosen CNN models. Extensive data augmentation techniques were employed to enhance the dataset’s diversity and generalization capabilities. We evaluated the models considering their accuracy, precision, recall, F1-score and computational efficiency to determine the most fitting model for real-life applications. The results demonstrate that all three CNN models achieved high performance in identifying and predicting potato leaf diseases, with VGG19 emerging as the top performer followed closely by DenseNet121 and ResNet50.Our findings provide valuable insights into the efficacy of DL approaches for potato leaf ailment identification and offer a foundation for future research and deployment of these models in precision agriculture systems. Ultimately, this work aims to support the development of more robust and efficient tools for timely disease diagnosis, enabling farmers and agronomists to make better-informed decisions and safeguard the health and productivity of potato crops worldwide.