Colon cancer prediction on histological images using deep learning features and Bayesian optimized SVM

Early and accurate identification of brain tumors is essential for determining effective treatment strategies and improving patient outcomes. Artificial intelligence (AI) and deep learning (DL) techniques have shown promise in automating diagnostic tasks based on magnetic resonance imaging (MRI). This study evaluates the performance of four pre-trained deep convolutional neural network (CNN) architectures for the automatic multi-class classification of brain tumors into four categories: Glioma, Meningioma, Pituitary, and No Tumor. The proposed approach utilizes the publicly accessible Brain Tumor MRI Msoud dataset, consisting of 7023 images, with 5712 provided for training and 1311 for testing. To assess the impact of data availability, subsets containing 25%, 50%, 75%, and 100% of the training data were used. A stratified five-fold cross-validation technique was applied. The CNN architectures evaluated include DeiT3_base_patch16_224, Xception41, Inception_v4, and Swin_Tiny_Patch4_Window7_224, all fine-tuned using transfer learning. The training pipeline incorporated advanced preprocessing and image data augmentation techniques to enhance robustness and mitigate overfitting. Among the models tested, Swin_Tiny_Patch4_Window7_224 achieved the highest classification Accuracy of 99.24% on the test set using 75% of the training data. This model demonstrated superior generalization across all tumor classes and effectively addressed class imbalance issues. Furthermore, we deployed and benchmarked the best-performing DL model on embedded AI platforms (Jetson AGX Xavier and Orin Nano), demonstrating their capability for real-time inference and highlighting their feasibility for edge-based clinical deployment. The results highlight the strong potential of pre-trained deep CNN and transformer-based architectures in medical image analysis. The proposed approach provides a scalable and energy-efficient solution for automated brain tumor diagnosis, facilitating the integration of AI into clinical workflows.

Colon cancer is cancer that is present on the inner side of colon walls or the rectum walls in the large intestine. Most of these types of cancer begin as abnormal growth of tissue called as polyp. Colonography uses low dose radiation Computed tomography (CT) scanning to obtain an interior view of the colon making use of special x-ray machine to view the large intestine for cancer and abnormal growths known as polyps. Radiologists examine these images to find polyp like structure using computer tools. As CT Colonography image contain noise such as lungs, small intestine, instruments during image capturing; segmenting colon from noise is the key task. Polyp occurrence can be detected mainly using shape feature; eliminating shapes similar to polyp is challenging. Hence, to tackle above issues, Image processing techniques are used by applying deep learning algorithm - Convolution Neural Network (CNN) and the results are compared with classical machine learning algorithm. In proposed method, each image is pre-processed to filter air filled dark region that includes colon, lungs etc. Next, each pre-processed CT Image separated into fixed number of blocks. Using pre-trained CNN, each block of ROI is classified as Type 1 (Usually Ascending and descending colon), Type 2 (Usually Traversal and sigmoidal colon) and Type3 (Noise such as lungs, instruments) to segment colon blocks by eliminating noise. Classified Blocks is further diagnosed for polyp like structure using pre-trained CNN by classifying each colon block as normal or abnormal. The experiment is setup with classical machine learning algorithms - Random Forest (RF) and k-nearest neighbor (KNN) by extracting texture feature - Local binary pattern (LBP) and shape feature - Histogram oriented gradient (HOG) for comparison. The experiment results showed the accuracy of proposed method for colon segmentation using CNN (87%) outperforms RF (85%) and KNN (83%). In, addition, the polyp detection accuracy of CNN (88%) is better than Random forest (85%) and KNN (80%). Hence, in the proposed method, there is significant accuracy improvement using deep learning algorithm compared to classical machine learning algorithms. It also provide baseline for automated colon cancer diagnosis using Deep learning algorithms for further research.

Here we investigate the application of convolutional neural networks (CNNs) to predict the survival time of patients with Glioblastoma Multiforme (GBM) brain tumor. Our dataset consists of T1-weighted high-resolution MRI images of just 68 GBM patients. We compare two analytic methods for predicting survival time. The first consists of training a small convolutional neural network (CNN) and the second uses extracted deep features from a pre-trained CNN. Our method is completely automated, except for tumor region segmentation. In addition, we utilize a snapshot ensemble approach to boost test accuracy when dealing with limited availability of medical images for CNN training purposes. Our approach achieves an accuracy of 72.06% using a trained small network and 66.18% using a pre-trained deep CNN. Our results compare favorably with the accuracy of 54.41% using histogram of oriented gradients (HOG) features and a non-neural network classifier.

Three convolutional neural network models for facial expression recognition in the wild

Comparative performance of four CNN-based deep learning variants in detecting Hispa pest, two fungal diseases, and NPK deficiency symptoms of rice (Oryza sativa)

zIn this research, a novel approach for classifying colon cancer was developed by employing two convolutional neural network (CNN) models, namely GoogLeNet and AlexNet.This approach involves training CNNs with histopathological images segmented into color clusters using an augmented k-means clustering algorithm, rather than utilizing original-raw images.This method was applied to 20 datasets with distinct structural and characteristic features, derived from larger datasets comprising both original and segmented images.The datasets were used to train and test CNN models.The results indicate that AlexNet, trained with segmented images, showed a 2% to 23% increase in accuracy performance, while GoogLeNet's accuracy performance improved by 2% to 27%.Notably, the proposed approach yielded higher accuracy with datasets containing non-homogeneous data. Bu aratrmada, GoogLeNet ve AlexNet olmak zere iki evriimli sinir a (CNN) modeli kullanlarak kolon kanserinin snflandrlmas iin yeni bir yaklam gelitirilmitir. Bu yaklamda CNN'ler, orijinal ham grntleri kullanmak yerine, artrlm bir k-ortalamalar kmeleme algoritmas kullanlarak renk kmelerine ayrlm histopatolojik grntleri kullanarak eitilmektedir. Bu yntem hem orijinal hem de bltlenmi grntlerden oluan daha byk veri kmelerinden elde edilen farkl yapsal ve karakteristik zelliklere sahip 20 veri kmesine uygulanmtr. Veri kmeleri CNN modellerini eitmek ve test etmek iin kullanlmtr. Sonular, blmlere ayrlm grntlerle eitilen AlexNet'in doruluk performansnda %2 ile %23 arasnda bir art gsterdiini, GoogLeNet

Recently Pretrained Convolutional Neural Networks (CNNs) have proven its effectiveness in image extraction and classification. This powerful feature of CNNs in image processing is facilitated by machine learning to train and classify big data. Image capturing and security transformation are considered as a central necessity of remote sensing imagery of unmanned aerial vehicles (UAVs) and drones. This paper presents a novel artificial intelligent drone-based encrypted machine learning of image classification using a pertained CNN and image encryption-decryption by utilizing singular value decomposition (SVD) and XOR-Secret-Key block cipher cryptology. Initially, pretrained convolutional neural networks (CNN) are extensively used to extract and classify image features making advantage of machine learning training tools features. Training of partial set of image data can be performed to test, classify and label the untrained image data. Pretrained CNN can classify images into object categories. Afterward, the CNN the classified image output is transformed into a digital matrix using SVD and identifies its associated eigenvalues. These eigenvalues are then converted into a binary code. The image data encryption is implemented according to suggested keys. The first part applies the exclusive OR (XOR) operation of the eigenvalues with a selected cipher key. Meanwhile, the second part implements the XOR operation of the output of part one with a randomly generated key using Poisson distribution. The last step in the encryption will be obtained by generating a non-real SVD decomposition matrix; according to which a non-readable image will be resulted. The original image-matrix can be constructed by reversing the process using the security key-cipher block (Poisson Distribution Key and Stand-alone Cipher Code). Finally, SVD image processing results are demonstrated to verify the effectiveness and security of the applied approach that can be implemented for different images.

In modern remote sensing image change detection (CD), convolution Neural Network (CNN), especially U-shaped structure (UNet), has achieved great success due to powerful discriminative ability. However, UNet-based CNN networks usually have limitations in modeling global dependencies due to intrinsic locality of convolution operations. Transformer has recently emerged as an alternative architecture for dense prediction tasks due to global self-attention mechanism. However, due to limitation of hardware resources, pure Transformer methods generally lack the ability to capture global information at a low level. Based on these existing problems, we propose STransUNet, which combines Transformer and UNet architecture. STransUNet can not only capture shallow detail features at an early stage, but also model global context in high-level feature. In addition, we design an efficient feature fusion module named Cross-Enhanced Adaptive Fusion(CEAF). Our model mainly consists of three parts: encoder, fusion module and decoder. The decoder is a CNN-Transformer hybrid structure. CNN extracts multi-level feature information. Transformer encodes tokenized sequence to capture global context. CEAF module cross-enhances and adaptively fuses bi-temporal features to enhance feature representation. In decoding stage, we introduce a Cascaded Upsampling decoder(CUP). CUP progressively aggregates low-level CNN features and high-level Transformer features to full resolution. On four public CD datasets, our STransUNet achieves better CD results than six state-of-the-art algorithms.

Deep learning based methods have recently been successfully explored in hyperspectral image classification field. However, training a deep learning model still requires a large number of labeled samples, which is usually impractical in hyperspectral images. In this paper, a simple but effective feature extraction method is proposed for hyperspectral image classification. Specifically, a pretrained deep convolutional neural network based on the ImageNet dataset is used to extract spatial features of a hyperspectral image. Recently, it is easy to obtain a pretrained convolutional neural network on the Internet. Note that the pretrained models are trained by using the ImageNet dataset. This means that the proposed method does not need labeled hyperspectral samples to train the deep model. Therefore, the proposed method alleviates the problem of lacking labeled samples and avoids the artificial design of feature extraction rules. Finally, the extracted features are stacked with spectral features as the input of a support vector machine classifier. The proposed method is conducted on three widely used hyperspectral image datasets. The experimental results demonstrate that the proposed method could outperform the conventional feature extraction methods and deep learning based methods.

IntroductionColon cancer (CC) image classification plays a key role in the diagnostic process in clinical contexts, especially as computational medical solutions become the trend for future radiology and pathology practices. This study presents a novel lightweight Convolutional Neural Network (CNN) model designed with effective data cleaning strategy for the classification and visualization of histopathology images of various colon cancer tissues.MethodsAddressing the critical need for efficient diagnostic tools in colon cancer detection, the proposed model leverages a non-pretrained architecture optimized for performance in resource-constrained environments. Utilizing the NCT-CRC-HE-100K and CRC-VAL-HE-7K datasets, this model employs a parametric Gaussian distribution-based data cleaning approach to enhance data quality by removing outliers.ResultsWith a total of 4,414,217 parameters and a total size of 16.9 megabytes, the model achieves a test accuracy of 0.990 ± 0.003 with 95% level of confidence, which demonstrates high precision, recall, specificity, and F1 scores across various tissue classes.DiscussionComparative analysis with benchmark studies underscores the model’s effectiveness, while discussions on underfitting and overfitting provide insights into potential fine-tuning strategies. This research presents a robust, lightweight solution for colon cancer histopathology image classification, offering a foundation for future advancements in colon cancer diagnostics with result visualization.

Precision medicine in oncology aims at obtaining data from heterogeneous sources to have a precise estimation of a given patient's state and prognosis. With the purpose of advancing to personalized medicine framework, accurate diagnoses allow prescription of more effective treatments adapted to the specificities of each individual case. In the last years, next-generation sequencing has impelled cancer research by providing physicians with an overwhelming amount of gene-expression data from RNA-seq high-throughput platforms. In this scenario, data mining and machine learning techniques have widely contribute to gene-expression data analysis by supplying computational models to supporting decision-making on real-world data. Nevertheless, existing public gene-expression databases are characterized by the unfavorable imbalance between the huge number of genes (in the order of tenths of thousands) and the small number of samples (in the order of a few hundreds) available. Despite diverse feature selection and extraction strategies have been traditionally applied to surpass derived over-fitting issues, the efficacy of standard machine learning pipelines is far from being satisfactory for the prediction of relevant clinical outcomes like follow-up end-points or patient's survival. Using the public Pan-Cancer dataset, in this study we pre-train convolutional neural network architectures for survival prediction on a subset composed of thousands of gene-expression samples from thirty-one tumor types. The resulting architectures are subsequently fine-tuned to predict lung cancer progression-free interval. The application of convolutional networks to gene-expression data has many limitations, derived from the unstructured nature of these data. In this work we propose a methodology to rearrange RNA-seq data by transforming RNA-seq samples into gene-expression images, from which convolutional networks can extract high-level features. As an additional objective, we investigate whether leveraging the information extracted from other tumor-type samples contributes to the extraction of high-level features that improve lung cancer progression prediction, compared to other machine learning approaches.

2.75D: Boosting learning by representing 3D Medical imaging to 2D features for small data

Recently, with the enormous growth of online videos, fast video retrieval research has received increasing attention. As an extension of image hashing techniques, traditional video hashing methods mainly depend on hand-crafted features and transform the real-valued features into binary hash codes. As videos provide far more diverse and complex visual information than images, extracting features from videos is much more challenging than that from images. Therefore, high-level semantic features to represent videos are needed rather than low-level hand-crafted methods. In this paper, a deep convolutional neural network is proposed to extract high-level semantic features and a binary hash function is then integrated into this framework to achieve an end-to-end optimization. Particularly, our approach also combines triplet loss function which preserves the relative similarity and difference of videos and classification loss function as the optimization objective. Experiments have been performed on two public datasets and the results demonstrate the superiority of our proposed method compared with other state-of-the-art video retrieval methods.

Purpose: The aim of the article is to develop an algorithm for classifying cracks in the analyzed images using modern methods of deep machine learning and transfer learning based on pretrained convolutional neural network - Inception-ResNet-v2. Design/Methodology/Approach: Transfer learning based on the pretrained convolutional neural network was used to categorize the images. The fully conected layer of the Inception-ResNet-v2 network has been modified. The last layer was trained using a two-class (binary) linear SVM (Support Vector Machine). In total, 20,000 training cases (images) were used to train the fully connected layer within transfer learning process. The research analyzed the possibility of using the deep neural networks for quick and fully automatic identification of cracks / defects on the surface of analyzed parts. Findings: The results indicate that pretrained convolutional neural network using SVM to train a fully connected layer is a very effective solution for visual crack / fault detection. In the analyzed model, a positive classification was obtained at the level of 99.89%. Practical Implications: The model presented in the article can be used in quality control carried out by process monitoring. An effective model for identifying defective parts can be used in both logistics and production processes. Originality/Value: A novelty is the use of a freely available, deep neural network trained to classify 1000 categories of various images for binary categorization of faults (cracks). The algorithm was adjusted by replacing the primary, 1000-output fully connected layer in the Inception-ResNet-v2 network with a binary layer (2 categories). The fully connected layer has been trained using the classification version of the popular SVM learner, but thanks to the combination of this layer with the sophisticated fearure extraction ability of the pre-trained Inception-ResNet-v2 deep network, the resulting predictive model enables the classification of defects with a very high level of accuracy.

Previous studies demonstrated that cancer sera contain antibodies, which react with a unique group of autologous cellular antigens called tumour-associated antigens (TAAs). This study determines whether a mini-array of multiple TAAs would enhance antibody detection and be a useful approach in colon cancer detection and diagnosis. The mini-array of multiple TAAs was composed of five TAAs including Imp1, p62, Koc, p53 and c-myc full-length recombinant proteins. Enzyme-linked immunosorbent assay (ELISA) was used to detect antibodies against these five TAAs in 46 sera from patients with colon cancer and also 58 sera from normal individuals. Antibody frequency to any individual TAA in colon cancer was variable and ranged from 15.2% to 23.9%. With the successive addition of TAAs to a final total of five antigens, there was a stepwise increase of positive antibody reactions reaching a sensitivity of 60.9% and a specificity of 89.7% in colon cancer. Positive and negative likelihood ratio was 5.91 and 0.43 respectively, which showed that the clinical diagnostic value of parallel assay of five TAAs was high. Positive and negative predictive values were respectively 82.4% and 74.3% indicating that parallel assay of five TAAs raised the diagnostic precision greatly. Agreement rate and Kappa value were 76.9% and 0.52 respectively, which indicated that the observed value of this assay had middle range coincidence with actual value. The data from this study further support our previous hypothesis that detection of autoantibodies for diagnosis of certain type of cancer can be enhanced by using a mini-array of several TAAs as target antigens. In 19 colon cancer sera with carcinoembryonic antigen (CEA) negative, 11 (57.9%) were found to have anti-TAA antibodies. When CEA and anti-TAAs were used together as markers in colon cancer detection, the diagnostic sensitivity could be raised from 60.9% to 82.6%. A customized antigen mini-array using a panel of appropriately selected TAAs can enhance autoantibody detection in immunodiagnosis of colon cancer. Anti-TAA and CEA were independent markers and the simultaneous use of these two markers significantly raised the sensitivity of colon cancer detection.