Computer-assisted Diagnosis Tools Research Articles

Computer-assisted urine cytology: Faster, cheaper, better?

Recent advancements in computer-assisted diagnosis (CAD) have catalysed significant progress in pathology, particularly in the realm of urine cytopathology. This review synthesizes the latest developments and challenges in CAD for diagnosing urothelial carcinomas, addressing the limitations of traditional urinary cytology. Through a literature review, we identify and analyse CAD models and algorithms developed for urine cytopathology, highlighting their methodologies and performance metrics. We discuss the potential of CAD to improve diagnostic accuracy, efficiency and patient outcomes, emphasizing its role in streamlining workflow and reducing errors. Furthermore, CAD tools have shown potential in exploring pathological conditions, uncovering novel biomarkers and prognostic/predictive features previously unknown or unseen. Finally, we examine the practical issues surrounding the integration of CAD into clinical practice, including regulatory approval, validation and training for pathologists. Despite the promising results, challenges remain, necessitating further research and validation efforts. Overall, CAD presents a transformative opportunity to revolutionize diagnostic practices in urine cytopathology, paving the way for enhanced patient care and outcomes.

A Space-Refine Paradigm for Automatic Carotid Artery Centerline Extraction in Magnetic Resonance Imaging.

Vessel centerline extraction is essential for carotid stenosis assessment and atherosclerotic plaque identification in clinical diagnosis. Simultaneously, it provides a region of interest identification and boundary initialization for computer-assisted diagnosis tools. In magnetic resonance imaging (MRI) cross-sectional images, the lumen shape and vascular topology result in a challenging task to extract the centerline accurately. To this end, we propose a space-refine framework, which exploits the positional continuity of the carotid artery from frame to frame to extract the carotid artery centerline. The proposed framework consists of a detector and a refinement module. Specifically, the detector roughly extracts the carotid lumen region from the original image. Then, we introduce a refinement module that uses the cascade of regressors from a detector to perform sequence realignment of lumen bounding boxes for each subject. It improves the lumen localization results and further enhances the centerline extraction accuracy. Verified by large carotid artery data, the proposed framework achieves state-of-the-art performance compared to conventional vessel centerline extraction methods or standard convolutional neural network approaches.Clinical relevance- Our proposed framework can be used as an important aid for physicians to quantitatively analyze the carotid artery in clinical practice. It is also used as a new paradigm for extracting the centerline of carotid vessels in computer-assisted tools.

Experiences Regarding Use and Implementation of Artificial Intelligence-Supported Follow-Up of Atypical Moles at a Dermatological Outpatient Clinic: Qualitative Study.

Artificial intelligence (AI) is increasingly used in numerous medical fields. In dermatology, AI can be used in the form of computer-assisted diagnosis (CAD) systems when assessing and diagnosing skin lesions suspicious of melanoma, a potentially lethal skin cancer with rising incidence all over the world. In particular, CAD may be a valuable tool in the follow-up of patients with high risk of developing melanoma, such as patients with multiple atypical moles. One such CAD system, ATBM Master (FotoFinder), can execute total body dermoscopy (TBD). This process comprises automatically photographing a patient´s entire body and then neatly displaying moles on a computer screen, grouped according to their clinical relevance. Proprietary FotoFinder algorithms underlie this organized presentation of moles. In addition, ATBM Master's optional convoluted neural network (CNN)-based Moleanalyzer Pro software can be used to further assess moles and estimate their probability of malignancy. Few qualitative studies have been conducted on the implementation of AI-supported procedures in dermatology. Therefore, the purpose of this study was to investigate how health care providers experience the use and implementation of a CAD system like ATBM Master, in particular its TBD module. In this way, the study aimed to elucidate potential barriers to the application of such new technology. We conducted a thematic analysis based on 2 focus group interviews with 14 doctors and nurses regularly working in an outpatient pigmented lesions clinic. Surprisingly, the study revealed that only 3 participants had actual experience using the TBD module. Even so, all participants were able to provide many notions and anticipations about its use, resulting in 3 major themes emerging from the interviews. First, several organizational matters were revealed to be a barrier to consistent use of the ATBM Master's TBD module, namely lack of guidance, time pressure, and insufficient training. Second, the study found that the perceived benefits of TBD were the ability to objectively detect and monitor subtle lesion changes and unbiasedness of the procedure. Imprecise identification of moles, inability to photograph certain areas, and substandard technical aspects were the perceived weaknesses. Lastly, the study found that clinicians were open to use AI-powered technology and that the TBD module was considered a supplementary tool to aid the medical staff, rather than a replacement of the clinician. Demonstrated by how few of the participants had actual experience with the TBD module, this study showed that implementation of new technology does not occur automatically. It highlights the importance of having a strategy for implementation to ensure the optimized application of CAD tools. The study identified areas that could be improved when implementing AI-powered technology, as well as providing insight on how medical staff anticipated and experienced the use of a CAD device in dermatology.

Deep Learning and Binary Relevance Classification of Multiple Diseases using Chest X-Ray images.

Disease detection using chest X-ray (CXR) images is one of the most popular radiology methods to diagnose diseases through a visual inspection of abnormal symptoms in the lung region. A wide variety of diseases such as pneumonia, heart failure and lung cancer can be detected using CXRs. Although CXRs can show the symptoms of a variety of diseases, detecting and manually classifying those diseases can be difficult and time-consuming adding to clinicians' work burden. Research shows that nearly 90% of mistakes made in a lung cancer diagnosis involved chest radiography. A variety of algorithms and computer-assisted diagnosis tools (CAD) were proposed to assist radiologists in the interpretation of medical images to reduce diagnosis errors. In this work, we propose a deep learning approach to screen multiple diseases using more than 220,000 images from the CheXpert dataset. The proposed binary relevance approach using Deep Convolutional Neural Networks (CNNs) achieves high performance results and outperforms past published work in this area.Clinical relevance- This application can be used to support physicians ans speed-up the diagnosis work. The proposed CAD can increase the confidence in the diagnosis or suggest a second opinion. The CAD can also be used in emergency situations when a radiologist is not available immediately.

A Computer-Assisted Diagnosis Tool for Classifying COVID-19 based on Chest X-Ray Images

Objective: Since COVID-19 is a worldwide pandemic, COVID-19 detection using a convolutional neural network (CNN) has been an extraordinary research technique. In the reported studies, many models that can predict COVID-19 based on deep learning methods using various medical images have been created; however, clinical decision support systems have been limited. The aim of this study is to develop a successful deep learning model based on X-ray images and a computer-assisted, fast, free and web-based diagnostic tool for accurate detection of COVID-19.
 Method: In this study a 15-layer CNN model was used to detect COVID-19 using X-ray images, which outperformed many previously published CNN models in terms of classification. The model performance is evaluated according to Accuracy, Matthews Correlation Coefficient (MCC), F1 Score, Specificity, Sensitivity (Recall), Youden’s Index, Precision (Positive Predictive Value: PPV), Negative Predictive Value (NPV), and Confusion Matrix (Classification matrix). In the second phase of the study, the computer-aided diagnostic tool for COVID-19 disease was developed using Python Flask library, JavaScript and Html codes.
 Results: The model to diagnose COVID-19 has an average accuracy of 98.68 % in the training set and 96.98 % in the testing set. Among the evaluation metrics, the minimum value is 93.4 % for MCC and Youden’s index, and the maximum value is 97.8 for sensitivity and NPV. A higher sensitivity value means a lower false negative (FN) value, and a low FN value is an encouraging outcome for COVID-19 cases. This conclusion is crucial because minimizing the overlooked cases of COVID-19 (false negatives) is one of the main goals of this research. 
 Conclusion: In this period when COVID-19 is spreading rapidly around the world, it is thought that the free and web-based COVID-19 X-Ray clinical decision support tool can be a very effective and fast diagnostic tool. The computer-aided system can assist physicians and radiologists in making clinical decisions about the disease, as well as provide support in diagnosis, follow-up, and prognosis. The developed computer-assisted diagnosis tool can be publicly accessed at http://biostatapps.inonu.edu.tr/CSYX/.

Deep-learning based detection of COVID-19 using lung ultrasound imagery

BackgroundThe COVID-19 pandemic has exposed the vulnerability of healthcare services worldwide, especially in underdeveloped countries. There is a clear need to develop novel computer-assisted diagnosis tools to provide rapid and cost-effective screening in places where massive traditional testing is not feasible. Lung ultrasound is a portable, easy to disinfect, low cost and non-invasive tool that can be used to identify lung diseases. Computer-assisted analysis of lung ultrasound imagery is a relatively recent approach that has shown great potential for diagnosing pulmonary conditions, being a viable alternative for screening and diagnosing COVID-19.ObjectiveTo evaluate and compare the performance of deep-learning techniques for detecting COVID-19 infections from lung ultrasound imagery.MethodsWe adapted different pre-trained deep learning architectures, including VGG19, InceptionV3, Xception, and ResNet50. We used the publicly available POCUS dataset comprising 3326 lung ultrasound frames of healthy, COVID-19, and pneumonia patients for training and fine-tuning. We conducted two experiments considering three classes (COVID-19, pneumonia, and healthy) and two classes (COVID-19 versus pneumonia and COVID-19 versus non-COVID-19) of predictive models. The obtained results were also compared with the POCOVID-net model. For performance evaluation, we calculated per-class classification metrics (Precision, Recall, and F1-score) and overall metrics (Accuracy, Balanced Accuracy, and Area Under the Receiver Operating Characteristic Curve). Lastly, we performed a statistical analysis of performance results using ANOVA and Friedman tests followed by post-hoc analysis using the Wilcoxon signed-rank test with the Holm’s step-down correction.ResultsInceptionV3 network achieved the best average accuracy (89.1%), balanced accuracy (89.3%), and area under the receiver operating curve (97.1%) for COVID-19 detection from bacterial pneumonia and healthy lung ultrasound data. The ANOVA and Friedman tests found statistically significant performance differences between models for accuracy, balanced accuracy and area under the receiver operating curve. Post-hoc analysis showed statistically significant differences between the performance obtained with the InceptionV3-based model and POCOVID-net, VGG19-, and ResNet50-based models. No statistically significant differences were found in the performance obtained with InceptionV3- and Xception-based models.ConclusionsDeep learning techniques for computer-assisted analysis of lung ultrasound imagery provide a promising avenue for COVID-19 screening and diagnosis. Particularly, we found that the InceptionV3 network provides the most promising predictive results from all AI-based techniques evaluated in this work. InceptionV3- and Xception-based models can be used to further develop a viable computer-assisted screening tool for COVID-19 based on ultrasound imagery.

Integrative Data Augmentation with U-Net Segmentation Masks Improves Detection of Lymph Node Metastases in Breast Cancer Patients.

Simple SummaryIn recent years many successful models have been developed to perform various tasks in digital histopathology, yet, there is still a reluctance to fully embrace the new technologies in clinical settings. One of the reasons for this is that although these models have achieved high performance at the patch-level, their performance at the image-level can still be underwhelming. Through this study, our main objective was to investigate whether integrating multiple extracted histological features to the input image had potential to further improve the performance of classifier models at the patch-level. Ideally, by achieving 100% accuracy at the patch-level, one can achieve 100% accuracy at the image-level. We hope that our research will entice the community to develop new strategies to further improve performance of existing state-of-the-art models, and facilitate their adoption in the clinics.Deep learning models have potential to improve performance of automated computer-assisted diagnosis tools in digital histopathology and reduce subjectivity. The main objective of this study was to further improve diagnostic potential of convolutional neural networks (CNNs) in detection of lymph node metastasis in breast cancer patients by integrative augmentation of input images with multiple segmentation channels. For this retrospective study, we used the PatchCamelyon dataset, consisting of 327,680 histopathology images of lymph node sections from breast cancer. Images had labels for the presence or absence of metastatic tissue. In addition, we used four separate histopathology datasets with annotations for nucleus, mitosis, tubule, and epithelium to train four instances of U-net. Then our baseline model was trained with and without additional segmentation channels and their performances were compared. Integrated gradient was used to visualize model attribution. The model trained with concatenation/integration of original input plus four additional segmentation channels, which we refer to as ConcatNet, was superior (AUC 0.924) compared to baseline with or without augmentations (AUC 0.854; 0.884). Baseline model trained with one additional segmentation channel showed intermediate performance (AUC 0.870-0.895). ConcatNet had sensitivity of 82.0% and specificity of 87.8%, which was an improvement in performance over the baseline (sensitivity of 74.6%; specificity of 80.4%). Integrated gradients showed that models trained with additional segmentation channels had improved focus on particular areas of the image containing aberrant cells. Augmenting images with additional segmentation channels improved baseline model performance as well as its ability to focus on discrete areas of the image.

A pap-smear analysis tool (PAT) for detection of cervical cancer from pap-smear images

BackgroundCervical cancer is preventable if effective screening measures are in place. Pap-smear is the commonest technique used for early screening and diagnosis of cervical cancer. However, the manual analysis of the pap-smears is error prone due to human mistake, moreover, the process is tedious and time-consuming. Hence, it is beneficial to develop a computer-assisted diagnosis tool to make the pap-smear test more accurate and reliable. This paper describes the development of a tool for automated diagnosis and classification of cervical cancer from pap-smear images.MethodScene segmentation was achieved through a Trainable Weka Segmentation classifier and a sequential elimination approach was used for debris rejection. Feature selection was achieved using simulated annealing integrated with a wrapper filter, while classification was achieved using a fuzzy C-means algorithm.ResultsThe evaluation of the classifier was carried out on three different datasets (single cell images, multiple cell images and pap-smear slide images from a pathology lab). Overall classification accuracy, sensitivity and specificity of ‘98.88%, 99.28% and 97.47%’, ‘97.64%, 98.08% and 97.16%’ and ‘95.00%, 100% and 90.00%’ were obtained for each dataset, respectively. The higher accuracy and sensitivity of the classifier was attributed to the robustness of the feature selection method that accurately selected cell features that improved the classification performance and the number of clusters used during defuzzification and classification. Results show that the method outperforms many of the existing algorithms in sensitivity (99.28%), specificity (97.47%), and accuracy (98.88%) when applied to the Herlev benchmark pap-smear dataset. False negative rate, false positive rate and classification error of 0.00%, 10.00% and 5.00%, respectively were obtained when applied to pap-smear slides from a pathology lab.ConclusionsThe major contribution of this tool in a cervical cancer screening workflow is that it reduces on the time required by the cytotechnician to screen very many pap-smears by eliminating the obvious normal ones, hence more time can be put on the suspicious slides. The proposed system has the capability of analyzing a full pap-smear slide within 3 min as opposed to the 5–10 min per slide in the manual analysis. The tool presented in this paper is applicable to many pap-smear analysis systems but is particularly pertinent to low-cost systems that should be of significant benefit to developing economies.

A neural network algorithm for detection of GI angiectasia during small-bowel capsule endoscopy

GI angiectasia (GIA) is the most common small-bowel (SB) vascular lesion, with an inherent risk of bleeding. SB capsule endoscopy (SB-CE) is the currently accepted diagnostic procedure. The aim of this study was to develop a computer-assisted diagnosis tool for the detection of GIA. Deidentified SB-CE still frames featuring annotated typical GIA and normal control still frames were selected from a database. A semantic segmentation images approach associated with a convolutional neural network (CNN) was used for deep-feature extractions and classification. Two datasets of still frames were created and used for machine learning and for algorithm testing. The GIA detection algorithm yielded a sensitivity of 100%, a specificity of 96%, a positive predictive value of 96%, and a negative predictive value of 100%. Reproducibility was optimal. The reading process for an entire SB-CE video would take 39 minutes. The developed CNN-based algorithm had high diagnostic performances, allowing detection of GIA in SB-CE still frames. This study paves the way for future automated CNN-based SB-CE reading softwares.

Domain-specific approach for segmentation of nucleus-cytoplasm in bone cancer histopathology for malignancy analysis

Bone cancer is prevalent and early detection of disease is need of the day. Most of the computer-assisted diagnosis tools of research pertain to various other parts of the body because of its simple tissue structure. Proposed research focuses on specific Ewing sarcoma stained with Haematoxylin and Eosin (H&E) data set wherein nucleus and cytoplasm features are extracted to define cancer. Relating to this two domain-specific segmentation algorithms are proposed: (1) model-based clustering, (2) gradient-based watershed segmentation methods are applied to extract nucleus and cytoplasm. H&E component of image are separated to improve quality of segmentation. Segmentation algorithms applied in parallel on H&E images to improve time factor. Morphological and texture features extracted from segmented images are used to train Support Vector Machine (SVM) for classification of malignancy.30 images of varying features are selected to train SVM classifier and 30 images are used for validation and the accuracy obtained is 94.5%.

On the development of conjunctival hyperemia computer-assisted diagnosis tools: Influence of feature selection and class imbalance in automatic gradings.

The sudden increase of blood flow in the bulbar conjunctiva, known as hyperemia, is associated to a red hue of variable intensity. Experts measure hyperemia using levels in a grading scale, a procedure that is subjective, non-repeatable and time consuming, thus creating a need for its automatisation. However, the task is far from straightforward due to data issues such as class imbalance or correlated features. In this paper, we study the specific features of hyperemia and propose various approaches to address these problems in the context of an automatic framework for hyperemia grading. Oversampling, undersampling and SMOTE approaches were applied in order to tackle the problem of class imbalance. 25 features were computed for each image and regression methods were then used to transform them into a value on the grading scale. The values and relationships among features and experts' values were analysed, and five feature selection techniques were subsequently studied. The lowest mean square error (MSE) for the regression systems trained with individual features is below 0.1 for both scales. Multi-layer perceptron (MLP) obtains the best values, but is less consistent than the random forest (RF) method. When all features are combined, the best results for both scales are achieved with MLP. Correlation based feature selection (CFS) and M5 provide the best results, MSE=0.108 and MSE=0.061 respectively. Finally, the class imbalance problem is minimised with the SMOTE approach for both scales (MSE<0.006). Machine learning methods are able to perform an objective assessment of hyperemia grading, removing both intra- and inter-expert subjectivity while providing a gain in computation time. SMOTE and oversampling approaches minimise the class imbalance problem, while feature selection reduces the number of features from 25 to 3-5 without worsening the MSE. As the differences between the system and a human expert are similar to the differences between experts, we can therefore conclude that the system behaves like an expert.

Representación de imágenes de histopatología utilizada en tareas de análisis automático: estado del arte

&lt;p&gt;This paper presents a review of the state-of-the-art in histopathology image representation used in automatic image analysis tasks. Automatic analysis of histopathology images is important for building computer-assisted diagnosis tools, automatic image enhancing systems and virtual microscopy systems, among other applications. Histopathology images have a rich mix of visual patterns with particularities that make them difficult to analyze. The paper discusses these particularities, the acquisition process and the challenges found when doing automatic analysis. Second an overview of recent works and methods addressed to deal with visual content representation in different automatic image analysis tasks is presented. Third an overview of applications of image representation methods in several medical domains and tasks is presented. Finally, the paper concludes with current trends of automatic analysis of histopathology images like digital pathology.&lt;/p&gt;

Alzheimer's diagnosis using eigenbrains and support vector machines

An accurate and early diagnosis of the Alzheimer&apos;s disease (AD) is of fundamental importance for the patient&apos;s medical treatment. Single photon emission computed tomography (SPECT) images are commonly used by physicians to assist the diagnosis. Presented is a computer-assisted diagnosis tool based in a principal component analysis (PCA) dimensional reduction of the feature space approach and a support vector machine (SVM) classification method for improving the AD diagnosis accuracy by means of SPECT images. The most relevant image features were selected under a PCA compression, which diagonalises the covariance matrix, and the extracted information was used to train an SVM classifier, which could classify new subjects in an unsupervised manner.