Fusing compressed deep ConvNets with a self-normalizing residual block and alpha dropout for a cost-efficient classification and diagnosis of gastrointestinal tract diseases

Montalbo, Francis Jesmar P.

doi:10.1016/j.mex.2022.101925

Cited by 5 publications

(2 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, Montalbo (2022) introduced a deep learning-based lightweight and cost-efficient state-of-the-art (SOTA) method utilizing KVASIR and ETIS-Larib Polyp DB datasets. The proposed approach seamlessly integrates network compression, layer-wise fusion, and the incorporation of a customized residual layer, denoted as the Modified Residual Block (MResBlock).…”

Section: Introductionmentioning

confidence: 99%

Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis

Hossain,

Shamrat,

Zhou

et al. 2024

PeerJ Computer Science

View full text Add to dashboard Cite

Gastrointestinal (GI) diseases are prevalent medical conditions that require accurate and timely diagnosis for effective treatment. To address this, we developed the Multi-Fusion Convolutional Neural Network (MF-CNN), a deep learning framework that strategically integrates and adapts elements from six deep learning models, enhancing feature extraction and classification of GI diseases from endoscopic images. The MF-CNN architecture leverages truncated and partially frozen layers from existing models, augmented with novel components such as Auxiliary Fusing Layers (AuxFL), Fusion Residual Block (FuRB), and Alpha Dropouts (αDO) to improve precision and robustness. This design facilitates the precise identification of conditions such as ulcerative colitis, polyps, esophagitis, and healthy colons. Our methodology involved preprocessing endoscopic images sourced from open databases, including KVASIR and ETIS-Larib Polyp DB, using adaptive histogram equalization (AHE) to enhance their quality. The MF-CNN framework supports detailed feature mapping for improved interpretability of the model’s internal workings. An ablation study was conducted to validate the contribution of each component, demonstrating that the integration of AuxFL, αDO, and FuRB played a crucial part in reducing overfitting and efficiency saturation and enhancing overall model performance. The MF-CNN demonstrated outstanding performance in terms of efficacy, achieving an accuracy rate of 99.25%. It also excelled in other key performance metrics with a precision of 99.27%, a recall of 99.25%, and an F1-score of 99.25%. These metrics confirmed the model’s proficiency in accurate classification and its capability to minimize false positives and negatives across all tested GI disease categories. Furthermore, the AUC values were exceptional, averaging 1.00 for both test and validation sets, indicating perfect discriminative ability. The findings of the P-R curve analysis and confusion matrix further confirmed the robust classification performance of the MF-CNN. This research introduces a technique for medical imaging that can potentially transform diagnostics in gastrointestinal healthcare facilities worldwide.

show abstract

Section: Introductionmentioning

confidence: 99%

Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis

Hossain,

Shamrat,

Zhou

et al. 2024

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…These endoscopic examinations even improve the analysis of the clinical characteristics of lesions for determining their type and severity and making proper diagnoses [5]. Differences in the knowledge of medical practitioner cause errors in certain cases, particularly with regard to problematic aspects of videos and images from endoscopy [6,7]. These inconsistencies may result in negative impact and misdiagnoses on patient care.…”

mentioning

confidence: 99%

Modified Salp Swarm Algorithm With Deep Learning Based Gastrointestinal Tract Disease Classification on Endoscopic Images

et al. 2023

View full text Add to dashboard Cite

Nowadays, the analysis of gastrointestinal (GI) tract disease utilzing endoscopic image classification becomes an active research activity from the biomedical sector. The latest technology in medical imaging is Wireless Capsule Endoscopy (WCE) for diagnosing gastrointestinal diseases namely bleeding, ulcer, polyp, and so on. Manual diagnoses will be time taking and tough for the medical practitioner; thus, the authors have designed computerized approaches for classifying and detecting such diseases. Many research groups presented various machine learning (ML) and image processing methods for classifying GI tract diseases in recent times. Conventional data augmentation and image processing methods are integrated with adjusted pre-trained deep convolutional neural networks (CNNs) for classifying diseases in the GI tract from WCI images. This study presents a Modified Salp Swarm Algorithm with Deep Learning based Gastrointestinal Tract Disease Classification (MSSADL-GITDC) on Endoscopic Images. The presented MSSADL-GITDC technique mainly focuses on the examination of WCE images for GIT classification. To accomplish this, the presented MSSADL-GITDC technique applies median filtering (MF) technique for image smoothening. The presented MSSADL-GITDC technique designs improved capsule network (CapsNet) model for feature extraction where the CapsNet model is modified by the class attention layer (CAL). Moreover, MSSA based hyperparameter tuning process is performed to improve the efficiency of the improved CapsNet model. For GIT classification, deep belief network with extreme learning machine (DBN-ELM) was used. Finally, backpropagation is applied for supervised fine tuning of the DBN-ELM model. The experimental validation of the MSSADL-GITDC technique takes place on Kvasir-V2 database reported the betterment of the MSSADL-GITDC technique on GIT classification with maximum accuracy of 98.03%.

show abstract

Gastro-CNN-VIT: Vision Transformer and Deep CNNs for Detecting GI Diseases in WCE Images

El Khalfaoui,

Alibouch,

El Ouafdi

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Fusing compressed deep ConvNets with a self-normalizing residual block and alpha dropout for a cost-efficient classification and diagnosis of gastrointestinal tract diseases

Cited by 5 publications

References 28 publications

Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis

Development of a multi-fusion convolutional neural network (MF-CNN) for enhanced gastrointestinal disease diagnosis in endoscopy image analysis

Modified Salp Swarm Algorithm With Deep Learning Based Gastrointestinal Tract Disease Classification on Endoscopic Images

Gastro-CNN-VIT: Vision Transformer and Deep CNNs for Detecting GI Diseases in WCE Images

Contact Info

Product

Resources

About