BTNeXt: A Deep Learning Based Transfer Model for Brain Tumor Classification

Purpose: Segmentation of the vessel tree from retinal fundus images can be used to track changes in the retina and be an important first step in a diagnosis. Manual segmentation is a time-consuming process that is prone to error; effective and reliable automation can alleviate these problems but one of the difficulties is uneven image background, which may affect segmentation performance. Approach: We present a patch-based deep learning framework, based on a modified U-Net architecture, that automatically segments the retinal blood vessels from fundus images. In particular, we evaluate how various pre-processing techniques, images with either no processing, N4 bias field correction, contrast limited adaptive histogram equalization (CLAHE), or a combination of N4 and CLAHE, can compensate for uneven image background and impact final segmentation performance. Results: We achieved competitive results on three publicly available datasets as a benchmark for our comparisons of pre-processing techniques. In addition, we introduce Bayesian statistical testing, which indicates little practical difference ( ) between pre-processing methods apart from the sensitivity metric. In terms of sensitivity and pre-processing, the combination of N4 correction and CLAHE performs better in comparison to unprocessed and N4 pre-processing ( ); but compared to CLAHE alone, the differences are not significant ( to 0.88). Conclusions: We conclude that deep learning is an effective method for retinal vessel segmentation and that CLAHE pre-processing has the greatest positive impact on segmentation performance, with N4 correction helping only in images with extremely inhomogeneous background illumination.

ObjectivesTo assess a new application of artificial intelligence for real‐time detection of laryngeal squamous cell carcinoma (LSCC) in both white light (WL) and narrow‐band imaging (NBI) videolaryngoscopies based on the You‐Only‐Look‐Once (YOLO) deep learning convolutional neural network (CNN).Study DesignExperimental study with retrospective data.MethodsRecorded videos of LSCC were retrospectively collected from in‐office transnasal videoendoscopies and intraoperative rigid endoscopies. LSCC videoframes were extracted for training, validation, and testing of various YOLO models. Different techniques were used to enhance the image analysis: contrast limited adaptive histogram equalization, data augmentation techniques, and test time augmentation (TTA). The best‐performing model was used to assess the automatic detection of LSCC in six videolaryngoscopies.ResultsTwo hundred and nineteen patients were retrospectively enrolled. A total of 624 LSCC videoframes were extracted. The YOLO models were trained after random distribution of images into a training set (82.6%), validation set (8.2%), and testing set (9.2%). Among the various models, the ensemble algorithm (YOLOv5s with YOLOv5m—TTA) achieved the best LSCC detection results, with performance metrics in par with the results reported by other state‐of‐the‐art detection models: 0.66 Precision (positive predicted value), 0.62 Recall (sensitivity), and 0.63 mean Average Precision at 0.5 intersection over union. Tests on the six videolaryngoscopies demonstrated an average computation time per videoframe of 0.026 seconds. Three demonstration videos are provided.ConclusionThis study identified a suitable CNN model for LSCC detection in WL and NBI videolaryngoscopies. Detection performances are highly promising. The limited complexity and quick computational times for LSCC detection make this model ideal for real‐time processing.Level of Evidence3 Laryngoscope, 132:1798–1806, 2022

In the case of neonates, especially low birth weight preterm and high-risk infants, portable X-rays are frequently used. However, the image quality of portable X-rays is significantly lower compared to standard adult or pediatric X-rays, leading to considerable challenges in identifying abnormalities. Although attempts have been made to introduce deep learning to address these image quality issues, the poor quality of the images themselves hinders the training of deep learning models, further emphasizing the need for image enhancement. Additionally, since neonates have a high cell division rate and are highly sensitive to radiation, increasing radiation exposure to improve image quality is not a viable solution. Therefore, it is crucial to enhance image quality through preprocessing before training deep learning models. While various image enhancement methods have been proposed, Contrast Limited Adaptive Histogram Equalization (CLAHE) has been recognized as an effective technique for contrast-based image improvement. However, despite extensive research, the process of setting CLAHE’s hyperparameters still relies on a brute force, manual approach, making it inefficient. To address this issue, we propose a method called Bayesian Optimization CLAHE(BO-CLAHE), which leverages Bayesian optimization to automatically select the optimal hyperparameters for X-ray images used in diagnosing lung diseases in preterm and high-risk neonates. The images enhanced by BO-CLAHE demonstrated superior performance across several classification models, with particularly notable improvements in diagnosing Transient Tachypnea of the Newborn (TTN). This approach not only reduces radiation exposure but also contributes to the development of AI-based diagnostic tools, playing a crucial role in the early diagnosis and treatment of preterm and high-risk neonates.

(Aim) Breast cancer is the most common cancer in women and the second most common cancer worldwide. With the rapid advancement of deep learning, the early stages of breast cancer development can be accurately detected by radiologists with the help of artificial intelligence systems. (Method) Based on mammographic imaging, a mainstream clinical breast screening technique, we present a diagnostic system for accurate classification of breast abnormalities based on ResNet-50. To improve the proposed model, we created a new data augmentation framework called SCDA (Scaling and Contrast limited adaptive histogram equalization Data Augmentation). In its procedure, we first conduct the scaling operation to the original training set, followed by applying contrast limited adaptive histogram equalisation (CLAHE) to the scaled training set. By stacking the training set after SCDA with the original training set, we formed a new training set. The network trained by the augmented training set, was coined as ResNet-SCDA-50. Our system, which aims at a binary classification on mammographic images acquired from INbreast and MINI-MIAS, classifies masses, microcalcification as "abnormal", while normal regions are classified as "normal". (Results) We present the first attempt to use the image contrast enhancement method as the data augmentation method, resulting in an averaged 98.55 percent specificity and 92.83 percent sensitivity, which gives our best model an overall accuracy of 95.74 percent. (Conclusion) Our proposed method is effective in classifying breast abnormality.

Efficient detection of foliar diseases in cassava (Manihot esculenta) is essential for sustaining crop productivity and ensuring food security, particularly in regions vulnerable to environmental stress. However, accurate identification remains a challenge due to the widespread occurrence of diseases such as Cassava Mosaic Disease (CMD), Cassava Bacterial Blight (CBB), and Cassava Brown Streak Disease (CBSD), which continue to threaten cassava yields. This study addresses two major obstacles in cassava disease classification—uneven image quality and imbalanced class distribution—by implementing Contrast Limited Adaptive Histogram Equalization (CLAHE) and the Synthetic Minority Over-sampling Technique (SMOTE). A publicly available dataset from the Cassava Leaf Disease Classification competition on Kaggle was used, and two pretrained convolutional neural networks, EfficientNetV2B2 and DenseNet169, were fine-tuned through transfer learning. The images were resized, enhanced using CLAHE, and augmented before being split into training, validation, and test sets. Both models were trained for 10 epochs using identical configurations. Results indicate that EfficientNetV2B2 achieved higher classification accuracy (88.1%) than DenseNet169 (86.4%), with CLAHE contributing a 2–3% improvement in accuracy. While these results are slightly lower than those reported in previous studies employing extended training durations and advanced techniques such as focal loss, the lightweight approach presented here proves effective under computational constraints. The findings demonstrate the feasibility of developing scalable and resource-efficient disease detection systems, especially for mobile or edge devices. Future research should focus on longer training schedules, advanced loss functions, and validation using field-acquired images to further improve model performance in real-world agricultural settings.

The identification of COVID-19 using chest X-ray (CXR) images plays a critical role in managing the pandemic by providing a rapid, non-invasive, and accessible diagnostic tool. This study evaluates the impact of different image preprocessing techniques on the performance of deep learning models for COVID-19 classification based on COVID-19 Radiography Database, which includes 10,192 normal CXR images, 6012 lung opacity (non-COVID lung infection) images, and 1345 viral pneumonia images. Along with the images, corresponding lung masks are also included to aid in the segmentation and analysis of lung regions. Specifically, three convolutional neural network (CNN) models were developed, each using a distinct preprocessing method: Contrast Limited Adaptive Histogram Equalization (CLAHE), traditional histogram equalization, and no preprocessing. The results revealed that while the CLAHE-enhanced model achieved the highest training accuracy (93.26%) and demonstrated superior stability during training, it showed lower performance in the validation phase, with validation accuracy of 91.31%. In contrast, the model with no preprocessing, which exhibited slightly lower training accuracy (92.98%), outperformed the CLAHE model during validation, achieving the highest validation accuracy of 91.50% and the lowest validation loss. The histogram equalization model demonstrated performance similar to that of CLAHE but with slightly higher validation loss and accuracy compared to the unprocessed model. These findings suggest that while CLAHE excels in enhancing image details during training, it may lead to overfitting and reduced generalization ability. In contrast, the model without preprocessing showed the best generalization and stability, indicating that preprocessing techniques should be chosen carefully to balance feature enhancement with the need for generalization in real-world applications. This study underscores the importance of selecting appropriate image preprocessing techniques to enhance deep learning models' performance in medical image classification, particularly for COVID-19 detection. Histogram Equalization The results contribute to ongoing efforts to optimize diagnostic tools using AI and image processing.

Breast cancer remains one of the most common cancers amongst women globally. Early detection is crucial for improving survival rates. While mammography is widely used and an effective imaging technique, it can sometimes yield false positive or false negatives. Mammogram interpretation is highly operator-dependent, introducing variability and the potential for diagnostic errors. Additionally, mammographic images have limitations, such as low contrast in breast tissue and overlapping structures that can obscure lesions or mimic abnormalities. These limitations can lead to unnecessary biopsies or delayed diagnosis. These challenges highlight the needs for advanced and data driven diagnostic tools to support and enhance current screening practices.</p> <p>In recent years, machine learning has emerged as a transformative force in the health care industry offering innovative solution to complex medical challenges. One of the most promising applications of this technology is in the early and accurate detection of breast cancer particularly through the mammography image analysis. Machine learning algorithms, especially deep learning models, have shown significant potential in identifying patterns in mammography that may indicate abnormal cases. This study utilizes the power of deep learning particularly Convolutional Neural Networks (CNNs) to enhance the classification of breast ultrasound images.</p> <p>This research proposes a novel two-level classification approach. At the first level classification, the system distinguishes normal from abnormal cases. Next, the second level classification, it further classifies abnormal cases into benign or malignant tumors. To address challenges such as limited and imbalanced datasets, the study incorporates comprehensive preprocessing techniques including contrast enhancement via Contrast Limited Adaptive Histogram Equalization (CLAHE) and region of interest (ROI). The preprocessing step were evaluated separately and in combination to assess their individual contributions. Additionally, data augmentation is utilized to improve model generalization. To further enhance performance and robustness, multiple CNN architectures are combined using an ensemble mechanism. Soft voting and hard voting leverage the strengths of diverse CNNs reducing individual model biases and improving overall accuracy.</p> <p>The main contributions of this research are: (1) the design of a two-level classification system (2) the utilization of multiple preprocessing techniques to improve image quality, (3) the application of voting ensemble learning to improve predictive stability and accuracy, and (4) an extensive evaluation on the BUSI dataset. This breaks down the complex diagnostic task into smaller, manageable steps to allow for more accurate comparisons, analysis, and improvements at each stage.</p> <p>The first preprocessing step involved the application of CLAHE. In the first level classification (normal vs abnormal), performance of various models has been evaluated. The models achieved an 82.05% accuracy for AlexNet, 84.62% for DenseNet, 85.26% for Inception, 82.69% for MobileNet, 82.05% for ResNet and 82.69% for VGG16. The accuracy of soft voting is 96.20% and accuracy of hard voting 92.09%. In the second level classification (benign vs malignant), the models achieved an 75.00% accuracy for AlexNet, 80.47% for DenseNet, 35.94% for Inception, 41.41% for MobileNet, 35.94% for ResNet and 82.81% for VGG16. The accuracy of soft voting is 90.84% and hard voting is 85.76%.</p> <p>The second preprocessing step involved the ROI was implemented. This step helped isolate the regions containing potential abnormalities, such as tumors, while filtering out irrelevant areas. After the application of ROI (normal vs abnormal), the system evaluated the performance of various models. In the first level classification (normal vs abnormal), the models achieved an 82.69% accuracy for AlexNet, 82.69% for DenseNet, 83.97% for Inception, 87.18% for MobileNet, 82.69% for ResNet and 82.05% for VGG16. The accuracy of soft voting is 100.00% and hard voting is 98.08%. In the second level classification (benign vs malignant), the system evaluated the performance of various models. The models achieved an 75.00% accuracy for AlexNet, 80.47% for DenseNet, 35.94% for Inception, 41.41% for MobileNet, 35.94% for ResNet and 82.81% for VGG16. Soft voting achieved 89.15% and hard voting achieved 80.95%.</p> <p>The final step in the preprocessing stage combines both CLAHE and ROI. The combination led to an improvement in classification performance. In the first level classification (normal vs abnormal), the models achieved an 91.77% accuracy for AlexNet, 92.72% for DenseNet, 95.57% for Inception, 96.52% for MobileNet, 76.90% for ResNet and 91.14% for VGG16. The accuracy of soft voting is 90.88% and hard voting is 93.99%. In the second level classification (benign vs malignant), the system evaluated the performance of various models. The models achieved an 87.69% accuracy for AlexNet, 99.23% for DenseNet, 96.92% for Inception, 93.08% for MobileNet, 94.62% for ResNet and 97.69% for VGG16. Soft voting achieved 96.92% and hard voting achieved 97.69%.</p> <p>These results indicate that every preprocessing step contributed to an improvement in classification accuracy across all models. This study demonstrates that a two-level, ensemble-based deep learning system can significantly improve the diagnostic accuracy of breast ultrasound images, offering a promising tool to assist radiologists and support early detection efforts, particularly in resource-constrained healthcare environments.

Deep-lymph: An advanced deep learning framework for precision diagnosis of lymphoma from histopathological images

Brain tumor is generally structured by the subtle summation of anomalous cells because it categorized as a mass of tissue, and it is very crucial to categorize brain tumors from Magnetic Resonance Imaging (MRI) for diagnosis purpose. The investigation of brain tumors by humans is a routine process for detecting and classifying MRI brain tumors. Due to the low contrast of the images, poor boundaries and noises, the image classification techniques provide poor performance in medical images. Hence, a new classification model for detecting brain tumors is needed to solve the critical challenges in existing approaches. This decreases human intervention while making decisions about brain tumors. Therefore, a new brain tumor classification model is investigated to solve the challenges in existing approaches with deep learning. The proposed framework consists of two main different phases (i) classifying normal and abnormal images and (b) Segmentation and severity classification of abnormal images. The first step is comprised of pre-processing and classification. Firstly, the required images are gathered from the benchmark databases. These collected images are undergone pre-processing, which is done by filtering methods and Contrast Limited Adaptive Histogram Equalization (CLAHE). Subsequently, the pre-processed image is fed into the new classification model as a Transformer-based Multi-scale Residual Attention Network (TMRAN), where the normal and abnormal images are obtained. While in the second case, the classified abnormal image is considered as the input for the segmentation process. The segmented image is obtained by Convolutional Neural Network (CNN), Modified UNet and UNet, in turn the parameters in UNet is optimized by using the Modified Controlling Parameters-based African Vultures Optimization Algorithm (MCP-AVOA), termed as Modified UNet. At the final stage, these segmented images are applied as input to the novel TMRAN model, where the hyper parameters are tuned optimally using improved MCP-AVOA in order to acquire the optimal severity classified results. The performance is validated using diverse measures and compared over other conventional methodologies. Thus, the findings elucidate that the proposed model attains higher classification results.

Adaptive histogram equalization (AHE) is a method for adaptive contrast enhancement of digital images. It is an automatic, reproducible method for the simultaneous viewing of contrast within a digital image with a large dynamic range. Recent experiments have shown that in specific cases, there is no significant difference in the ability of AHE and linear intensity windowing to display gray-scale contrast. More recently, a variant of AHE which limits the allowed contrast enhancement of the image has been proposed. This contrast-limited adaptive histogram equalization (CLAHE) produces images in which the noise content of an image is not excessively enhanced, but in which sufficient contrast is provided for the visualization of structures within the image. Images processed with CLAHE have a more natural appearance and facilitate the comparison of different areas of an image. However, the reduced contrast enhancement of CLAHE may hinder the ability of an observer to detect the presence of some significant gray-scale contrast. In this report, a psychophysical observer experiment was performed to determine if there is a significant difference in the ability of AHE and CLAHE to depict gray-scale contrast. Observers were presented with computed tomography (CT) images of the chest processed with AHE and CLAHE. Subtle artificial lesions were introduced into some images. The observers were asked to rate their confidence regarding the presence of the lesions; this rating-scale data was analyzed using receiver operating characteristic (ROC) curve techniques. These ROC curves were compared for significant differences in the observers' performances. In this report, no difference was found in the abilities of AHE and CLAHE to depict contrast information.

Across the globe, dyslexia and dysgraphia are two frequent learning disorders identified in classrooms. This condition is characterized by difficulties in age-appropriate reading without any sociocultural restrictions. Children with this disorder have difficulty recognizing word and letter patterns. Early identification of dyslexic children (DC) is crucial for providing them with the most effective educational opportunities. Researchers proposed a deep learning-based dyslexia detection system (DDS). However, there is a demand for a practical, lightweight framework for identifying DC. Thus, the proposed study intends to build a framework for detecting dyslexia. The proposed framework encompasses image processing, feature extraction, and classification models. The image-processing model enhances the image quality using contrast-limited adaptive histogram equalization and resizes the images into 512 × 512 pixels. For feature extraction, the authors employ you only look once V7 to extract features in a limited time. In addition, the MobileNet V2 with single shot detection lite is used to classify the handwritten images into normal and abnormal classes, respectively. The authors utilized the publicly available dyslexia dataset for performance evaluation. The test set contains 19,557 normal and 17,882 reversal (abnormal) images. The baseline models are employed for comparative analysis. The experimental study revealed that the proposed framework outperformed the baseline models by achieving exceptional precision, recall, F1-Score, accuracy, and mean average precision of 97.9, 97.3, 97.6, 99.2, and 97.6, respectively. In addition, the proposed model obtained an exceptional mean intersection over union of 88.6. It can be implemented in educational institutions and healthcare centers. In the future, the authors can extend the research to build an integrated framework using biomedical images.

The most common causes of blindness in adults worldwide, 2.6% of them are caused by diabetic retinopathy, which is a progressive disease caused by complications of diabetes mellitus. Early detection and prompt treatment help save eyesight. The artificial intelligence technology can provide objective and accurate screening results. Deep learning, especially the Convolutional Neural Network (CNN), which is part of artificial intelligence, has proven successful in solving image problems and is very well used in medical image analysis. CNN works well on large datasets, but it will affect network performance to overfitting in fewer datasets. So to solve the problem of small data, data augmentation techniques can be used. There are various kinds of data augmentation techniques. This study used the CNN method to classify diabetic retinopathy disease, compared several suitable data augmentation techniques for retinal fundus images, and used Contrast Limited Adaptive Histogram Equalization (CLAHE) for image enhancement. This study found that the augmented random zoom technique, together with CLAHE, provided the best accuracy of 98% with 96% sensitivity and 100% specificity.

Vision loss can be avoided if diabetic retinopathy (DR) is diagnosed and treated promptly. The main five DR stages are none, moderate, mild, proliferate, and severe. In this study, a deep learning (DL) model is presented that diagnoses all five stages of DR with more accuracy than previous methods. The suggested method presents two scenarios: case 1 with image enhancement using a contrast limited adaptive histogram equalization (CLAHE) filtering algorithm in conjunction with an enhanced super-resolution generative adversarial network (ESRGAN), and case 2 without image enhancement. Augmentation techniques were then performed to generate a balanced dataset utilizing the same parameters for both cases. Using Inception-V3 applied to the Asia Pacific Tele-Ophthalmology Society (APTOS) datasets, the developed model achieved an accuracy of 98.7% for case 1 and 80.87% for case 2, which is greater than existing methods for detecting the five stages of DR. It was demonstrated that using CLAHE and ESRGAN improves a model’s performance and learning ability.

This study aims to develop a deep learning-based model for accurate fruit disease classification using advanced architectures like ResNet50 and VGG19. The dataset includes images of fruits such as grapes, mangoes, lemons, pomegranates, and guavas, with both healthy and diseased categories. Diseases like black rot in grapes and bacterial canker in mangoes are considered for classification. The objective is to build a model capable of accurately identifying these diseases to aid in early identification and control. The approach begins with pre-processing the photographs using standardisation, resizing, and data augmentation techniques including flipping, zooming, and rotation to strengthen the model. By enhancing picture contrast with Contrast Limited Adaptive Histogram Equalisation (CLAHE), important illness characteristics can be better highlighted. We train and evaluate deep learning models by dividing the dataset into three parts: training, validation, and test. The ratio of the parts is 80:10:10. According on the results, the ResNet50 model outperforms all other accuracy at 99.77%.

Accurate detection of egg fertility is crucial to improve hatching success in duck farming. Conventional candling methods rely heavily on human expertise, making them subjective and error-prone. This study proposes an automated classification system for Cihateup duck egg fertility using candling images and a convolutional neural network (CNN) based on the EfficientNet-B3 architecture. Image enhancement techniques, including contrast limited adaptive histogram equalization (CLAHE), unsharp masking, and adaptive thresholding, were applied to improve image quality and feature visibility. The dataset consisted of fertile and infertile egg images captured at two incubation stages: the first 24 hours and the 8th–15th days. Data were split into training, validation, and testing sets with a ratio of 70:15:15. Experimental results show that image enhancement significantly improves classification performance. Without enhancement, the model achieved an accuracy of 49% with an area under curve (AUC) of 0.4226, indicating poor discrimination capability. With image enhancement, the proposed method achieved accuracies of 77% for the first 24 hours dataset and 80% for the 8th–15th days dataset, with AUC values of 0.9962 and 0.9317, respectively. These results demonstrate that EfficientNet-B3 combined with image enhancement provides an effective and computationally efficient solution for automated fertility detection of Cihateup duck eggs.