This report describes 12 major practices during data handling that predispose machine learning systems to mathematical bias. Minimizing bias is critical to adoption and implementation of machine learning (ML) in clinical practice. Systematic mathematical biases produce consistent and reproducible differences between the ...

Machine learning, and especially deep learning, is rapidly gaining acceptance and clinical usage in a wide range of image analysis applications and is regarded as providing high performance in detecting anatomical structures and identification and classification of patterns of disease in medical images. However, there are many roadblocks to the widespread implementation of machine learning in clinical image analysis, including differences in data capture leading to different measurements, high dimensionality of imaging and other medical data, and the black-box nature of machine learning, with a lack of insight into relevant features. Techniques such as radiomics have been used in traditional machine learning approaches to model the mathematical relationships between adjacent pixels in an image and provide an explainable framework for clinicians and researchers. Newer paradigms, such as topological data analysis (TDA), have recently been adopted to design and develop innovative image analysis schemes that go beyond the abilities of pixel-to-pixel comparisons. TDA can automatically construct filtrations of topological shapes of image texture through a technique known as persistent homology (PH); these features can then be fed into machine learning models that provide explainable outputs and can distinguish different image classes in a computationally more efficient way, when compared to other currently used methods. The aim of this review is to introduce PH and its variants and to review TDA’s recent successes in medical imaging studies.

“Just Accepted” papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review ...

Parkinson’s disease (PD) is a devastating neurological disease that cannot be identified with traditional plasma experiments, necessitating the development of a faster, less expensive diagnostic instrument. Due to the difficulty of quantifying PD in the past, doctors have tended to focus on some signs while ignoring others, primarily relying on an intuitive assessment scale because of the disease’s characteristics, which include loss of motor control and speech that can be utilized to detect and diagnose this disease. It is an illness that impacts both motion and non-motion functions. It takes years to develop and has a wide range of clinical symptoms and prognoses. Parkinson’s patients commonly display non-motor symptoms such as sleep problems, neurocognitive ailments, and cognitive impairment long before the diagnosis, even though scientists have been working to develop designs for diagnosing and categorizing the disease, only noticeable defects such as movement patterns, speech, or writing skills are offered in this paper. This article provides a thorough analysis of several AI-based ML and DL techniques used to diagnose PD and their influence on developing additional research directions. It follows the guidelines of Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR). This review also examines the current state of PD diagnosis and the potential applications of data-driven AI technology. It ends with a discussion of future developments, which aids in filling critical gaps in the current Parkinson’s study.

This report reviews potential biases during machine learning model development and proposes possible mitigation strategies for radiology artificial intelligence studies. There are increasing concerns about the bias and fairness of artificial intelligence (AI) models as they are put into clinical practice. Among the steps for implementing machine learning tools into...