Bone diseases are common and can result in various musculoskeletal conditions (MC). An estimated 1.71 billion patients suffer from musculoskeletal problems worldwide. Apart from musculoskeletal fractures, femoral neck injuries, knee osteoarthritis, and fractures are very common bone diseases, and the rate is expected to double in the next 30 years. Therefore, proper and timely diagnosis and treatment of a fractured patient are crucial. Contrastingly, missed fractures are a common prognosis failure in accidents and emergencies. This causes complications and delays in patients’ treatment and care. These days, artificial intelligence (AI) and, more specifically, deep learning (DL) are receiving significant attention to assist radiologists in bone fracture detection. DL can be widely used in medical image analysis. Some studies in traumatology and orthopaedics have shown the use and potential of DL in diagnosing fractures and diseases from radiographs. In this systematic review, we provide an overview of the use of DL in bone imaging to help radiologists to detect various abnormalities, particularly fractures. We have also discussed the challenges and problems faced in the DL-based method, and the future of DL in bone imaging.

The global healthcare sector continues to grow rapidly and is reflected as one of the fastest-growing sectors in the fourth industrial revolution (4.0). The majority of the healthcare industry still uses labor-intensive, time-consuming, and error-prone traditional, manual, and manpower-based methods. This review addresses the current paradigm, the potential for new scientific discoveries, the technological state of preparation, the potential for supervised machine learning (SML) prospects in various healthcare sectors, and ethical issues. The effectiveness and potential for innovation of disease diagnosis, personalized medicine, clinical trials, non-invasive image analysis, drug discovery, patient care services, remote patient monitoring, hospital data, and nanotechnology in various learning-based automation in healthcare along with the requirement for explainable artificial intelligence (AI) in healthcare are evaluated. In order to understand the potential architecture of non-invasive treatment, a thorough study of medical imaging analysis from a technical point of view is presented. This study also represents new thinking and developments that will push the boundaries and increase the opportunity for healthcare through AI and SML in the near future. Nowadays, SML-based applications require a lot of data quality awareness as healthcare is data-heavy, and knowledge management is paramount. Nowadays, SML in biomedical and healthcare developments needs skills, quality data consciousness for data-intensive study, and a knowledge-centric health management system. As a result, the merits, demerits, and precautions need to take ethics and the other effects of AI and SML into consideration. The overall insight in this paper will help researchers in academia and industry to understand and address the future research that needs to be discussed on SML in the healthcare and biomedical sectors.

An entity’s existence in an image can be depicted by the activity instantiation vector from a group of neurons (called capsule). Recently, multi-layered capsules, called CapsNet, have proven to be state-of-the-art for image classification tasks. This research utilizes the prowess of this algorithm to detect pneumonia from chest X-ray (CXR) images. Here, an entity in the CXR image can help determine if the patient (whose CXR is used) is suffering from pneumonia or not. A simple model of capsules (also known as Simple CapsNet) has provided results comparable to best Deep Learning models that had been used earlier. Subsequently, a combination of convolutions and capsules is used to obtain two models that outperform all models previously proposed. These models—Integration of convolutions with capsules (ICC) and Ensemble of convolutions with capsules (ECC)—detect pneumonia with a test accuracy of 95.33% and 95.90%, respectively. The latter model is studied in detail to obtain a variant called EnCC, where n = 3, 4, 8, 16. Here, the E4CC model works optimally and gives test accuracy of 96.36%. All these models had been trained, validated, and tested on 5857 images from Mendeley.

Radiomics analyses has been proposed to interrogate the biology of tumour as well as to predict/assess response to therapy in vivo. The objective of this work was to assess the sensitivity of radiomics features to noise, resolution, and tumour volume in the context of a co-clinical trial. Triple negative breast cancer (TNBC) patients were recruited into an ongoing co-clinical imaging trial. Sub-typed matched TNBC patient-derived tumour xenografts (PDX) were generated to investigate optimal co-clinical MR radiomic features. The MR imaging protocol included T1-weighed and T2-weighted imaging. To test the sensitivity of radiomics to resolution, PDX were imaged at three different resolutions. Multiple sets of images with varying signal-to-noise ratio (SNR) were generated, and an image independent patch-based method was implemented to measure the noise levels. Forty-eight radiomic features were extracted from manually segmented 2D and 3D segmented tumours and normal tissues of T1- and T2- weighted co-clinical MR images. Sixteen radiomics features were identified as volume dependent and corrected for volume-dependency following normalization. Features from grey-level run-length matrix (GLRLM), grey-level size zone matrix (GLSZM) were identified as most sensitive to noise. Radiomic features Kurtosis and Run-length variance (RLV) from GLSZM were most sensitive to changes in resolution in both T1w and T2w MRI. In general, 3D radiomic features were more robust compared to 2D (single slice) measures, although the former exhibited higher variability between subjects. Tumour volume, noise characteristics, and image resolution significantly impact radiomic analysis in co-clinical studies.