Knee osteoarthritis severity classification with ordinal regression module

MedKnee: A New Deep Learning-Based Software for Automated Prediction of Radiographic Knee Osteoarthritis

In computer-aided medical diagnosis, deep learning techniques have shown that it is possible to offer performance similar to that of experienced medical specialists in the diagnosis of knee osteoarthritis. In this study, a new deep learning (DL) software, called "MedKnee" is developed to assist physicians in the diagnosis process of knee osteoarthritis according to the Kellgren and Lawrence (KL) score. To accomplish this task, 5000 knee X-ray images obtained from the Osteoarthritis Initiative public dataset (OAI) were divided into train, valid, and test datasets in a ratio of 7:1:2 with a balanced distribution across each KL grade. The pre-trained Xception model is used for transfer learning and then deployed in a Graphical User Interface (GUI) developed with Tkinter and Python. The suggested software was validated on an external public database, Medical Expert, and compared with a rheumatologist's diagnosis on a local database, with the involvement of a radiologist for arbitration. The MedKnee achieved an accuracy of 95.36% when tested on Medical Expert-I and 94.94% on Medical Expert-II. In the local dataset, the developed tool and the rheumatologist agreed on 23 images out of 30 images (74%). The MedKnee's satisfactory performance makes it an effective assistant for doctors in the assessment of knee osteoarthritis.

Ensemble deep-learning networks for automated osteoarthritis grading in knee X-ray images

The Kellgren-Lawrence (KL) grading system is a scoring system for classifying the severity of knee osteoarthritis using X-ray images, and it is the standard X-ray-based grading system for diagnosing knee osteoarthritis. However, KL grading depends on the clinician's subjective assessment. Moreover, the accuracy varies significantly depending on the clinician's experience and can be particularly low. Therefore, in this study, we developed an ensemble network that can predict a consistent and accurate KL grade for knee osteoarthritis severity using a deep learning approach. We trained individual models on knee X-ray datasets using the most suitable image size for each model in an ensemble network rather than using datasets with a single image size. We then built the ensemble network using these models to overcome the instability of single models and further improve accuracy. We conducted various experiments using a dataset of 8260 images from the Osteoarthritis Initiative open dataset. The proposed ensemble network exhibited the best performance, achieving an accuracy of 76.93% and an F1-score of 0.7665. The Grad-CAM visualization technique was used to further evaluate the focus of the model. The results demonstrated that the proposed ensemble network outperforms existing techniques that have performed well in KL grade classification. Moreover, the proposed model focuses on the joint space around the knee to extract the imaging features required for KL grade classification, revealing its high potential for diagnosing knee osteoarthritis.

The Efficacy and Safety of Intra-articular Low Molecular Weight Fraction of Human Serum Albumin for the Management of Moderate to Moderately Severe Knee Osteoarthritis: A Systematic Review and Meta-Analysis

Osteoarthritis is a chronic degenerative joint disease that affects weight-bearing joints. Low molecular weight fraction of 5% (LMWF-5A) human serum albumin is an intra-articular injection that emerged for the treatment of knee osteoarthritis. The aim of this review is to assess the efficacy and safety of LMWF-5A versus placebo through a systematic review and meta-analysis. The Cochrane Central Register of Controlled Trials (CENTRAL), Medical Literature Analysis and Retrieval System Online (MEDLINE), EBSCO, and ClinicalTrials.gov registry databases were utilized to search for studies. Only randomized controlled trials (RCTs) that evaluated the efficacy of LMWF-5A versus placebo were included. Efficacy endpoints were represented by Western Ontario and McMaster Universities Arthritis Index (WOMAC) A and C scores for pain and function, respectively. Serious adverse events (SAEs), non-serious adverse events (NSAEs), and mortality rates were used to evaluate the safety of the drug. The revised Cochrane risk of bias tool was used for the risk of bias assessment. Seven RCTs (n=2939) that met the inclusion criteria were included. The meta-analysis did not find significant improvement in pain (WOMAC A) (standardized mean difference (SMD)= -0.01, 95% confidence interval (CI) -0.10 - 0.09, P=0.87, I²=30%). Additionally, no significant change in function was noted (WOMAC C) (SMD=0.01, 95% CI -0.08 - 0.10, P=0.87, I²=22%). The pooled analysis did not find a significant difference between LMWF-5A and placebo regarding the incidence of joint swelling (P=0.84), joint stiffness (P=0.53), arthralgia (P=0.53), extremity pain (P=0.45), NSAEs (P=0.21), SAEs (P=0.92), or mortality (P=1.00). However, the subgroup analysis showed a significant reduction of 42% in NSAEs upon administration of 10 mL of LMWF-5A (risk ratio (RR)=0.58, 95% CI 0.35-0.97, P=0.04). In summary, our meta-analysis did not find significant differences between LMWF-5A and placebo regarding the incidence of NSAEs, SAEs, or mortality. On the other hand, LMWF-5A did not demonstrate superiority over saline in terms of efficacy. Therefore, it is not an effective drug for managing knee osteoarthritis.

Knee Osteoarthritis Detection and Severity Classification Using Residual Neural Networks on Preprocessed X-ray Images

One of the most common and challenging medical conditions to deal with in old-aged people is the occurrence of knee osteoarthritis (KOA). Manual diagnosis of this disease involves observing X-ray images of the knee area and classifying it under five grades using the Kellgren-Lawrence (KL) system. This requires the physician's expertise, suitable experience, and a lot of time, and even after that the diagnosis can be prone to errors. Therefore, researchers in the ML/DL domain have employed the capabilities of deep neural network (DNN) models to identify and classify KOA images in an automated, faster, and accurate manner. To this end, we propose the application of six pretrained DNN models, namely, VGG16, VGG19, ResNet101, MobileNetV2, InceptionResNetV2, and DenseNet121 for KOA diagnosis using images obtained from the Osteoarthritis Initiative (OAI) dataset. More specifically, we perform two types of classification, namely, a binary classification, which detects the presence or absence of KOA and secondly, classifying the severity of KOA in a three-class classification. For a comparative analysis, we experiment on three datasets (Dataset I, Dataset II, and Dataset III) with five, two, and three classes of KOA images, respectively. We achieved maximum classification accuracies of 69%, 83%, and 89%, respectively, with the ResNet101 DNN model. Our results show an improved performance from the existing work in the literature.

A fully automatic target detection and quantification strategy based on object detection convolutional neural network YOLOv3 for one-step X-ray image grading

Methods for automatic image analysis are demanded for dealing with the explosively increased imaging data in clinics. Osteoarthritis (OA) is a typical disease diagnosed based on X-ray imaging. Herein, we propose a novel modeling strategy based on YOLO version 3 (YOLOv3) for automatic simultaneous localization of knee joints and quantification of radiographic knee OA. As an advanced deep convolutional neural network (CNN) algorithm for target detection, YOLOv3 enables simultaneous small object detection and quantification due to its unique residual connection and feature map merging. Hence, a unified CNN model is built for the elegant integration of knee joint detection and corresponding OA severity grading using the YOLOv3 framework. We achieve desirable accuracy in knee OA grading using the public and clinical datasets. It provides improvements in the precision, recall, F₁ score and diagnostic accuracy of knee OA as well. Because of the fully automatic target detection and quantification, the time of handling an image is merely 40 ms from inputting the image to getting its label, supporting quick clinic decisions. It, thus, affords convenient and efficient image analysis for daily clinical diagnosis.

Transfer learning-assisted 3D deep learning models for knee osteoarthritis detection: Data from the osteoarthritis initiative

Knee osteoarthritis is one of the most common musculoskeletal diseases and is usually diagnosed with medical imaging techniques. Conventionally, case identification using plain radiography is practiced. However, we acknowledge that knee osteoarthritis is a 3D complexity; hence, magnetic resonance imaging will be the ideal modality to reveal the hidden osteoarthritis features from a three-dimensional view. In this work, the feasibility of well-known convolutional neural network (CNN) structures (ResNet, DenseNet, VGG, and AlexNet) to distinguish knees with and without osteoarthritis (OA) is investigated. Using 3D convolutional layers, we demonstrated the potential of 3D convolutional neural networks of 13 different architectures in knee osteoarthritis diagnosis. We used transfer learning by transforming 2D pre-trained weights into 3D as initial weights for the training of the 3D models. The performance of the models was compared and evaluated based on the performance metrics [balanced accuracy, precision, F1 score, and area under receiver operating characteristic (AUC) curve]. This study suggested that transfer learning indeed enhanced the performance of the models, especially for ResNet and DenseNet models. Transfer learning-based models presented promising results, with ResNet34 achieving the best overall accuracy of 0.875 and an F1 score of 0.871. The results also showed that shallow networks yielded better performance than deeper neural networks, demonstrated by ResNet18, DenseNet121, and VGG11 with AUC values of 0.945, 0.914, and 0.928, respectively. This encourages the application of clinical diagnostic aid for knee osteoarthritis using 3DCNN even in limited hardware conditions.

Discovering Knee Osteoarthritis Imaging Features for Diagnosis and Prognosis: Review of Manual Imaging Grading and Machine Learning Approaches

Knee osteoarthritis (OA) is a deliberating joint disorder characterized by cartilage loss that can be captured by imaging modalities and translated into imaging features. Observing imaging features is a well-known objective assessment for knee OA disorder. However, the variety of imaging features is rarely discussed. This study reviews knee OA imaging features with respect to different imaging modalities for traditional OA diagnosis and updates recent image-based machine learning approaches for knee OA diagnosis and prognosis. Although most studies recognized X-ray as standard imaging option for knee OA diagnosis, the imaging features are limited to bony changes and less sensitive to short-term OA changes. Researchers have recommended the usage of MRI to study the hidden OA-related radiomic features in soft tissues and bony structures. Furthermore, ultrasound imaging features should be explored to make it more feasible for point-of-care diagnosis. Traditional knee OA diagnosis mainly relies on manual interpretation of medical images based on the Kellgren-Lawrence (KL) grading scheme, but this approach is consistently prone to human resource and time constraints and less effective for OA prevention. Recent studies revealed the capability of machine learning approaches in automating knee OA diagnosis and prognosis, through three major tasks: knee joint localization (detection and segmentation), classification of OA severity, and prediction of disease progression. AI-aided diagnostic models improved the quality of knee OA diagnosis significantly in terms of time taken, reproducibility, and accuracy. Prognostic ability was demonstrated by several prediction models in terms of estimating possible OA onset, OA deterioration, progressive pain, progressive structural change, progressive structural change with pain, and time to total knee replacement (TKR) incidence. Despite research gaps, machine learning techniques still manifest huge potential to work on demanding tasks such as early knee OA detection and estimation of future disease events, as well as fundamental tasks such as discovering the new imaging features and establishment of novel OA status measure. Continuous machine learning model enhancement may favour the discovery of new OA treatment in future.

Prioritisation Assessment and Robust Predictive System for Medical Equipment: A Comprehensive Strategic Maintenance Management

The advancement of technology in medical equipment has significantly improved healthcare services. However, failures in upkeeping reliability, availability, and safety affect the healthcare services quality and significant impact can be observed in operations' expenses. The effective and comprehensive medical equipment assessment and monitoring throughout the maintenance phase of the asset life cycle can enhance the equipment reliability, availability, and safety. The study aims to develop the prioritisation assessment and predictive systems that measure the priority of medical equipment's preventive maintenance, corrective maintenance, and replacement programmes. The proposed predictive model is constructed by analysing features of 13,352 medical equipment used in public healthcare clinics in Malaysia. The proposed system comprises three stages: prioritisation analysis, model training, and predictive model development. In this study, we proposed 16 combinations of novel features to be used for prioritisation assessment and prediction of preventive maintenance, corrective maintenance, and replacement programme. The modified k-Means algorithm is proposed during the prioritisation analysis to automatically distinguish raw data into three main clusters of prioritisation assessment. Subsequently, these clusters are fed into and tested with six machine learning algorithms for the predictive prioritisation system. The best predictive models for medical equipment's preventive maintenance, corrective maintenance, and replacement programmes are selected among the tested machine learning algorithms. Findings indicate that the Support Vector Machine performs the best in preventive maintenance and replacement programme prioritisation predictive systems with the highest accuracy of 99.42 and 99.80%, respectively. Meanwhile, K-Nearest Neighbour yielded the highest accuracy in corrective maintenance prioritisation predictive systems with 98.93%. Based on the promising results, clinical engineers and healthcare providers can widely adopt the proposed prioritisation assessment and predictive systems in managing expenses, reporting, scheduling, materials, and workforce.

An ensemble learning method based on ordinal regression for COVID-19 diagnosis from chest CT

Coronavirus disease 2019 (COVID-19) has brought huge losses to the world, and it remains a great threat to public health. X-ray computed tomography (CT) plays a central role in the management of COVID-19. Traditional diagnosis with pulmonary CT images is time-consuming and error-prone, which could not meet the need for precise and rapid COVID-19 screening. Nowadays, deep learning (DL) has been successfully applied to CT image analysis, which assists radiologists in workflow scheduling and treatment planning for patients with COVID-19. Traditional methods use cross-entropy as the loss function with a Softmax classifier following a fully-connected layer. Most DL-based classification methods target intraclass relationships in a certain class (early, progressive, severe, or dissipative phases), ignoring the natural order of different phases of the disease progression,i.e.,from an early stage and progress to a late stage. To learn both intraclass and interclass relationships among different stages and improve the accuracy of classification, this paper proposes an ensemble learning method based on ordinal regression, which leverages the ordinal information on COVID-19 phases. The proposed method uses multi-binary, neuron stick-breaking (NSB), and soft labels (SL) techniques, and ensembles the ordinal outputs through a median selection. To evaluate our method, we collected 172 confirmed cases. In a 2-fold cross-validation experiment, the accuracy is increased by 22% compared with traditional methods when we use modified ResNet-18 as the backbone. And precision, recall, andF1-score are also improved. The experimental results show that our proposed method achieves a better classification performance than the traditional methods, which helps establish guidelines for the classification of COVID-19 chest CT images.

Vibrational Spectroscopy in Assessment of Early Osteoarthritis-A Narrative Review

Osteoarthritis (OA) is a degenerative disease, and there is currently no effective medicine to cure it. Early prevention and treatment can effectively reduce the pain of OA patients and save costs. Therefore, it is necessary to diagnose OA at an early stage. There are various diagnostic methods for OA, but the methods applied to early diagnosis are limited. Ordinary optical diagnosis is confined to the surface, while laboratory tests, such as rheumatoid factor inspection and physical arthritis checks, are too trivial or time-consuming. Evidently, there is an urgent need to develop a rapid nondestructive detection method for the early diagnosis of OA. Vibrational spectroscopy is a rapid and nondestructive technique that has attracted much attention. In this review, near-infrared (NIR), infrared, (IR) and Raman spectroscopy were introduced to show their potential in early OA diagnosis. The basic principles were discussed first, and then the research progress to date was discussed, as well as its limitations and the direction of development. Finally, all methods were compared, and vibrational spectroscopy was demonstrated that it could be used as a promising tool for early OA diagnosis. This review provides theoretical support for the application and development of vibrational spectroscopy technology in OA diagnosis, providing a new strategy for the nondestructive and rapid diagnosis of arthritis and promoting the development and clinical application of a component-based molecular spectrum detection technology.