Adapting content-based image retrieval techniques for the semantic annotation of medical images

Artificial intelligence model for tumoral clinical decision support systems

BACKGROUND AND OBJECTIVE

Comparative diagnostic in brain tumor evaluation makes possible to use the available information of a medical center to compare similar cases when a new patient is evaluated. By leveraging Artificial Intelligence models, the proposed system is able of retrieving the most similar cases of brain tumors for a given query. The primary objective is to enhance the diagnostic process by generating more accurate representations of medical images, with a particular focus on patient-specific normal features and pathologies. A key distinction from previous models lies in its ability to produce enriched image descriptors solely from binary information, eliminating the need for costly and difficult to obtain tumor segmentation.

METHODS

The proposed model uses Artificial Intelligence to detect patient features to recommend the most similar cases from a database. The system not only suggests similar cases but also balances the representation of healthy and abnormal features in its design. This not only encourages the generalization of its use but also aids clinicians in their decision-making processes. This generalization makes possible for future research in different medical diagnosis areas with almost not any change in the system.

RESULTS

We conducted a comparative analysis of our approach in relation to similar studies. The proposed architecture obtains a Dice coefficient of 0.474 in both tumoral and healthy regions of the patients, which outperforms previous literature. Our proposed model excels at extracting and combining anatomical and pathological features from brain Magnetic Resonances (MRs), achieving state-of-the-art results while relying on less expensive label information. This substantially reduces the overall cost of the training process. Our findings highlight the significant potential for improving the efficiency and accuracy of comparative diagnostics and the treatment of tumoral pathologies.

CONCLUSIONS

This paper provides substantial grounds for further exploration of the broader applicability and optimization of the proposed architecture to enhance clinical decision-making. The novel approach presented in this work marks a significant advancement in the field of medical diagnosis, particularly in the context of Artificial Intelligence-assisted image retrieval, and promises to reduce costs and improve the quality of patient care using Artificial Intelligence as a support tool instead of a black box system.

Optimizing vitiligo diagnosis with ResNet and Swin transformer deep learning models: a study on performance and interpretability

Vitiligo is a hypopigmented skin disease characterized by the loss of melanin. The progressive nature and widespread incidence of vitiligo necessitate timely and accurate detection. Usually, a single diagnostic test often falls short of providing definitive confirmation of the condition, necessitating the assessment by dermatologists who specialize in vitiligo. However, the current scarcity of such specialized medical professionals presents a significant challenge. To mitigate this issue and enhance diagnostic accuracy, it is essential to build deep learning models that can support and expedite the detection process. This study endeavors to establish a deep learning framework to enhance the diagnostic accuracy of vitiligo. To this end, a comparative analysis of five models including ResNet (ResNet34, ResNet50, and ResNet101 models) and Swin Transformer series (Swin Transformer Base, and Swin Transformer Large models), were conducted under the uniform condition to identify the model with superior classification capabilities. Moreover, the study sought to augment the interpretability of these models by selecting one that not only provides accurate diagnostic outcomes but also offers visual cues highlighting the regions pertinent to vitiligo. The empirical findings reveal that the Swin Transformer Large model achieved the best performance in classification, whose AUC, accuracy, sensitivity, and specificity are 0.94, 93.82%, 94.02%, and 93.5%, respectively. In terms of interpretability, the highlighted regions in the class activation map correspond to the lesion regions of the vitiligo images, which shows that it effectively indicates the specific category regions associated with the decision-making of dermatological diagnosis. Additionally, the visualization of feature maps generated in the middle layer of the deep learning model provides insights into the internal mechanisms of the model, which is valuable for improving the interpretability of the model, tuning performance, and enhancing clinical applicability. The outcomes of this study underscore the significant potential of deep learning models to revolutionize medical diagnosis by improving diagnostic accuracy and operational efficiency. The research highlights the necessity for ongoing exploration in this domain to fully leverage the capabilities of deep learning technologies in medical diagnostics.

Performance and clinical applicability of machine learning in liver computed tomography imaging: a systematic review

OBJECTIVES

Machine learning (ML) for medical imaging is emerging for several organs and image modalities. Our objectives were to provide clinicians with an overview of this field by answering the following questions: (1) How is ML applied in liver computed tomography (CT) imaging? (2) How well do ML systems perform in liver CT imaging? (3) What are the clinical applications of ML in liver CT imaging?

METHODS

A systematic review was carried out according to the guidelines from the PRISMA-P statement. The search string focused on studies containing content relating to artificial intelligence, liver, and computed tomography.

RESULTS

One hundred ninety-one studies were included in the study. ML was applied to CT liver imaging by image analysis without clinicians' intervention in majority of studies while in newer studies the fusion of ML method with clinical intervention have been identified. Several were documented to perform very accurately on reliable but small data. Most models identified were deep learning-based, mainly using convolutional neural networks. Potentially many clinical applications of ML to CT liver imaging have been identified through our review including liver and its lesion segmentation and classification, segmentation of vascular structure inside the liver, fibrosis and cirrhosis staging, metastasis prediction, and evaluation of chemotherapy.

CONCLUSION

Several studies attempted to provide transparent result of the model. To make the model convenient for a clinical application, prospective clinical validation studies are in urgent call. Computer scientists and engineers should seek to cooperate with health professionals to ensure this.

KEY POINTS

• ML shows great potential for CT liver image tasks such as pixel-wise segmentation and classification of liver and liver lesions, fibrosis staging, metastasis prediction, and retrieval of relevant liver lesions from similar cases of other patients. • Despite presenting the result is not standardized, many studies have attempted to provide transparent results to interpret the machine learning method performance in the literature. • Prospective studies are in urgent call for clinical validation of ML method, preferably carried out by cooperation between clinicians and computer scientists.

Improving the Automatic Classification of Brain MRI Acquisition Contrast with Machine Learning

Automated quantification of data acquired as part of an MRI exam requires identification of the specific acquisition of relevance to a particular analysis. This motivates the development of methods capable of reliably classifying MRI acquisitions according to their nominal contrast type, e.g., T1 weighted, T1 post-contrast, T2 weighted, T2-weighted FLAIR, proton-density weighted. Prior studies have investigated using imaging-based methods and DICOM metadata-based methods with success on cohorts of patients acquired as part of a clinical trial. This study compares the performance of these methods on heterogeneous clinical datasets acquired with many different scanners from many institutions. RF and CNN models were trained on metadata and pixel data, respectively. A combined RF model incorporated CNN logits from the pixel-based model together with metadata. Four cohorts were used for model development and evaluation: MS research (n = 11,106 series), MS clinical (n = 3244 series), glioma research (n = 612 series, test/validation only), and ADNI PTSD (n = 477 series, training only). Together, these cohorts represent a broad range of acquisition contexts (scanners, sequences, institutions) and subject pathologies. Pixel-based CNN and combined models achieved accuracies between 97 and 98% on the clinical MS cohort. Validation/test accuracies with the glioma cohort were 99.7% (metadata only) and 98.4 (CNN). Accurate and generalizable classification of MRI acquisition contrast types was demonstrated. Such methods are important for enabling automated data selection in high-throughput and big-data image analysis applications.

Automatic Generation of Structured Radiology Reports for Volumetric Computed Tomography Images Using Question-Specific Deep Feature Extraction and Learning

BACKGROUND

In today's modern medicine, the use of radiological imaging devices has spread at medical centers. Therefore, the need for accurate, reliable, and portable medical image analysis and understanding systems has been increasing constantly. Accompanying images with the required clinical information, in the form of structured reports, is very important, because images play a pivotal role in detect, planning, and diagnosis of different diseases. Report-writing can be exposure to error, tedious and labor-intensive for physicians and radiologists; to address these issues, there is a need for systems that generate medical image reports automatically and efficiently. Thus, automatic report generation systems are among the most desired applications.

METHODS

This research proposes an automatic structured-radiology report generation system that is based on deep learning methods. Extracting useful and descriptive image features to model the conceptual contents of the images is one of the main challenges in this regard. Considering the ability of deep neural networks (DNNs) in soliciting informative and effective features as well as lower resource requirements, tailored convolutional neural networks and MobileNets are employed as the main building blocks of the proposed system. To cope with challenges such as multi-slice medical images and diversity of questions asked in a radiology report, our system develops volume-level and question-specific deep features using DNNs.

RESULTS

We demonstrate the effectiveness of the proposed system on ImageCLEF2015 Liver computed tomography (CT) annotation task, for filling in a structured radiology report about liver CT. The results confirm the efficiency of the proposed approach, as compared to classic annotation methods.

CONCLUSION

We have proposed a question-specific DNNbased system for filling in structured radiology reports about medical images.

A Visually Interpretable Deep Learning Framework for Histopathological Image-Based Skin Cancer Diagnosis

Owing to the high incidence rate and the severe impact of skin cancer, the precise diagnosis of malignant skin tumors is a significant goal, especially considering treatment is normally effective if the tumor is detected early. Limited published histopathological image sets and the lack of an intuitive correspondence between the features of lesion areas and a certain type of skin cancer pose a challenge to the establishment of high-quality and interpretable computer-aided diagnostic (CAD) systems. To solve this problem, a light-weight attention mechanism-based deep learning framework, namely, DRANet, is proposed to differentiate 11 types of skin diseases based on a real histopathological image set collected by us during the last 10 years. The CAD system can output not only the name of a certain disease but also a visualized diagnostic report showing possible areas related to the disease. The experimental results demonstrate that the DRANet obtains significantly better performance than baseline models (i.e., InceptionV3, ResNet50, VGG16, and VGG19) with comparable parameter size and competitive accuracy with fewer model parameters. Visualized results produced by the hidden layers of the DRANet actually highlight part of the class-specific regions of diagnostic points and are valuable for decision making in the diagnosis of skin diseases.

Deep metric learning-based image retrieval system for chest radiograph and its clinical applications in COVID-19

In recent years, deep learning-based image analysis methods have been widely applied in computer-aided detection, diagnosis and prognosis, and has shown its value during the public health crisis of the novel coronavirus disease 2019 (COVID-19) pandemic. Chest radiograph (CXR) has been playing a crucial role in COVID-19 patient triaging, diagnosing and monitoring, particularly in the United States. Considering the mixed and unspecific signals in CXR, an image retrieval model of CXR that provides both similar images and associated clinical information can be more clinically meaningful than a direct image diagnostic model. In this work we develop a novel CXR image retrieval model based on deep metric learning. Unlike traditional diagnostic models which aim at learning the direct mapping from images to labels, the proposed model aims at learning the optimized embedding space of images, where images with the same labels and similar contents are pulled together. The proposed model utilizes multi-similarity loss with hard-mining sampling strategy and attention mechanism to learn the optimized embedding space, and provides similar images, the visualizations of disease-related attention maps and useful clinical information to assist clinical decisions. The model is trained and validated on an international multi-site COVID-19 dataset collected from 3 different sources. Experimental results of COVID-19 image retrieval and diagnosis tasks show that the proposed model can serve as a robust solution for CXR analysis and patient management for COVID-19. The model is also tested on its transferability on a different clinical decision support task for COVID-19, where the pre-trained model is applied to extract image features from a new dataset without any further training. The extracted features are then combined with COVID-19 patient's vitals, lab tests and medical histories to predict the possibility of airway intubation in 72 hours, which is strongly associated with patient prognosis, and is crucial for patient care and hospital resource planning. These results demonstrate our deep metric learning based image retrieval model is highly efficient in the CXR retrieval, diagnosis and prognosis, and thus has great clinical value for the treatment and management of COVID-19 patients.

Employing Divergent Machine Learning Classifiers to Upgrade the Preciseness of Image Retrieval Systems

Content Based Image Retrieval (CBIR) system is an efficient search engine which has the potentiality of retrieving the images from huge repositories by extracting the visual features. It includes color, texture and shape. Texture is the most eminent feature among all. This investigation focuses upon the classification complications that crop up in case of big datasets. In this, texture techniques are explored with machine learning algorithms in order to increase the retrieval efficiency. We have tested our system on three texture techniques using various classifiers which are Support vector machine, K-Nearest Neighbor (KNN), Naïve Bayes and Decision Tree (DT). Variant evaluation metrics precision, recall, false alarm rate, accuracy etc. are figured out to measure the competence of the designed CBIR system on two benchmark datasets, i.e. Wang and Brodatz. Result shows that with both these datasets the KNN and DT classifier hand over superior results as compared to others.

Computational framework for fusing eye movements and spoken narratives for image annotation

Despite many recent advances in the field of computer vision, there remains a disconnect between how computers process images and how humans understand them. To begin to bridge this gap, we propose a framework that integrates human-elicited gaze and spoken language to label perceptually important regions in an image. Our work relies on the notion that gaze and spoken narratives can jointly model how humans inspect and analyze images. Using an unsupervised bitext alignment algorithm originally developed for machine translation, we create meaningful mappings between participants' eye movements over an image and their spoken descriptions of that image. The resulting multimodal alignments are then used to annotate image regions with linguistic labels. The accuracy of these labels exceeds that of baseline alignments obtained using purely temporal correspondence between fixations and words. We also find differences in system performances when identifying image regions using clustering methods that rely on gaze information rather than image features. The alignments produced by our framework can be used to create a database of low-level image features and high-level semantic annotations corresponding to perceptually important image regions. The framework can potentially be applied to any multimodal data stream and to any visual domain. To this end, we provide the research community with access to the computational framework.

OtoMatch: Content-based eardrum image retrieval using deep learning

Acute infections of the middle ear are the most commonly treated childhood diseases. Because complications affect children's language learning and cognitive processes, it is essential to diagnose these diseases in a timely and accurate manner. The prevailing literature suggests that it is difficult to accurately diagnose these infections, even for experienced ear, nose, and throat (ENT) physicians. Advanced care practitioners (e.g., nurse practitioners, physician assistants) serve as first-line providers in many primary care settings and may benefit from additional guidance to appropriately determine the diagnosis and treatment of ear diseases. For this purpose, we designed a content-based image retrieval (CBIR) system (called OtoMatch) for normal, middle ear effusion, and tympanostomy tube conditions, operating on eardrum images captured with a digital otoscope. We present a method that enables the conversion of any convolutional neural network (trained for classification) into an image retrieval model. As a proof of concept, we converted a pre-trained deep learning model into an image retrieval system. We accomplished this by changing the fully connected layers into lookup tables. A database of 454 labeled eardrum images (179 normal, 179 effusion, and 96 tube cases) was used to train and test the system. On a 10-fold cross validation, the proposed method resulted in an average accuracy of 80.58% (SD 5.37%), and maximum F1 score of 0.90 while retrieving the most similar image from the database. These are promising results for the first study to demonstrate the feasibility of developing a CBIR system for eardrum images using the newly proposed methodology.

Data Tagging in Medical Images: A Survey of the State-of-Art

A huge amount of medical data is generated every second, and a significant percentage of the data are images that need to be analyzed and processed. One of the key challenges in this regard is the recovery of the data of medical images. The medical image recovery procedure should be done automatically by the computers that are the method of identifying object concepts and assigning homologous tags to them. To discover the hidden concepts in the medical images, the lowlevel characteristics should be used to achieve high-level concepts and that is a challenging task. In any specific case, it requires human involvement to determine the significance of the image. To allow machine-based reasoning on the medical evidence collected, the data must be accompanied by additional interpretive semantics; a change from a pure data-intensive methodology to a model of evidence rich in semantics. In this state-of-art, data tagging methods related to medical images are surveyed which is an important aspect for the recognition of a huge number of medical images. Different types of tags related to the medical image, prerequisites of medical data tagging, different techniques to develop medical image tags, different medical image tagging algorithms and different tools that are used to create the tags are discussed in this paper. The aim of this state-of-art paper is to produce a summary and a set of guidelines for using the tags for the identification of medical images and to identify the challenges and future research directions of tagging medical images.

Ontologies for Liver Diseases Representation: A Systematic Literature Review

Ontology, as a useful knowledge engineering technique, has been widely used for reducing ambiguity and helping with information sharing. It is considered originally to be clear, comprehensive, and with well-defined format. It characterizes several domains purposes description through structured and formalized languages. In various areas of research, it has become a significant way to realize successful and powerful accomplishments. Actually, medical ontologies were turned into an efficient application in medical domains. They also become a relevant approach to process large medical data volumes. Consequently, they are behaving as a support decision system in some cases. Also, they ensure diagnosis process acceleration and assistance. Additionally, they have been integrated especially to represent human healthcare concepts. For that reason, plenty of research works applied ontologies to design and treat liver diseases. In this article, we present a general overview of medical ontologies to stand for this type of disease. We expose and discuss these works in details by a complete comparison. Also, we show their performance to arrange clinical data and extract results.

Generate Structured Radiology Report from CT Images Using Image Annotation Techniques: Preliminary Results with Liver CT

A medical annotation system for radiology images extracts clinically useful information from the images, allowing the machines to infer useful abstract semantics and become capable of automatic reasoning and making diagnostic decision. It also supplies human-interpretable explanation for the images. We have implemented a computerized framework that, given a liver CT image, predicts radiological annotations with high accuracy, in order to generate a structured report, which includes predicting very specific high-level semantic content. Each report of a liver CT image is related to different inhomogeneous parts like the liver, lesion, and vessel. We put forward a claim that gathering all kinds of features is not suitable for filling all parts of the report. As a matter of fact, for each group of annotations, one should find and extract the best feature that results in the best answers for that specific annotation. To this end, the main challenge is discovering the relationships between these specific semantic concepts and their association with the low-level image features. Our framework was implemented by combining a set of the state-of-the-art low-level imaging features. In addition, we propose a novel feature (DLBP (deep local binary pattern)) based on LBP that incorporates multi-slice analysis in CT images and further improves the performance. In order to model our annotation system, two methods were used, namely multi-class support vector machine (SVM) and random subspace (RS) which is an ensemble learning method. Applying this representation leads to a high prediction accuracy of 93.1% despite its relatively low dimension in comparison with the existing works.

An Automatic Classification Method on Chronic Venous Insufficiency Images

Chronic venous insufficiency (CVI) affect a large population, and it cannot heal without doctors' interventions. However, many patients do not get the medical advisory service in time. At the same time, the doctors also need an assistant tool to classify the patients according to the severity level of CVI. We propose an automatic classification method, named CVI-classifier to help doctors and patients. In this approach, first, low-level image features are mapped into middle-level semantic features by a concept classifier, and a multi-scale semantic model is constructed to form the image representation with rich semantics. Second, a scene classifier is trained using an optimized feature subset calculated by the high-order dependency based feature selection approach, and is used to estimate CVI's severity. At last, classification accuracy, kappa coefficient, F1-score are used to evaluate classification performance. Experiments on the CVI images from 217 patients' medical records demonstrated superior performance and efficiency for CVI-classifier, with classification accuracy up to 90.92%, kappa coefficient of 0.8735 and F1score of 0.9006. This method also outperformed doctors' diagnosis (doctors rely solely on images to make judgments) with accuracy, kappa and F1-score improved by 9.11%, 0.1250 and 0.0955 respectively.

A fully automatic end-to-end method for content-based image retrieval of CT scans with similar liver lesion annotations

PURPOSE

The goal of medical content-based image retrieval (M-CBIR) is to assist radiologists in the decision-making process by retrieving medical cases similar to a given image. One of the key interests of radiologists is lesions and their annotations, since the patient treatment depends on the lesion diagnosis. Therefore, a key feature of M-CBIR systems is the retrieval of scans with the most similar lesion annotations. To be of value, M-CBIR systems should be fully automatic to handle large case databases.

METHODS

We present a fully automatic end-to-end method for the retrieval of CT scans with similar liver lesion annotations. The input is a database of abdominal CT scans labeled with liver lesions, a query CT scan, and optionally one radiologist-specified lesion annotation of interest. The output is an ordered list of the database CT scans with the most similar liver lesion annotations. The method starts by automatically segmenting the liver in the scan. It then extracts a histogram-based features vector from the segmented region, learns the features' relative importance, and ranks the database scans according to the relative importance measure. The main advantages of our method are that it fully automates the end-to-end querying process, that it uses simple and efficient techniques that are scalable to large datasets, and that it produces quality retrieval results using an unannotated CT scan.

RESULTS

Our experimental results on 9 CT queries on a dataset of 41 volumetric CT scans from the 2014 Image CLEF Liver Annotation Task yield an average retrieval accuracy (Normalized Discounted Cumulative Gain index) of 0.77 and 0.84 without/with annotation, respectively.

CONCLUSIONS

Fully automatic end-to-end retrieval of similar cases based on image information alone, rather that on disease diagnosis, may help radiologists to better diagnose liver lesions.

Computer-Aided Medical Image Annotation: Preliminary Results With Liver Lesions in CT

The increasing volume of medical image data, as well as the need for multicenter data consolidation for big data analytics, require computer-aided medical image annotation (CMIA). Majority of the methods proposed so far do not exploit interdependencies between annotations explicitly. They further limit their annotations at a higher level than diagnostics and/or do not consider a standardized lexicon. A radiologist-in-the-loop semi-automatic CMIA system is proposed. It is based on a Bayesian tree structured model, linked to RadLex, and present preliminary results with liver lesions in computed tomography images. The proposed system guides the radiologist to input the most critical information in each iteration and uses a network model to update the full annotation online. The effectiveness of the system using this model-based interactive annotation scheme is shown by contrasting the domain-blind and domain-aware models. Preliminary results show that on average 7.50 (out of 29) manual annotations are sufficient for ${\text{95}}\%$ accuracy, which is ${\text{32.8}}\%$ less than the required manual effort when there is no guidance. The results also suggest that the domain-aware models perform better than the domain-blind models learned from data. Further analysis with larger datasets and in domains other than the liver lesions is needed.