Abstract:Medical imaging is a key component in clinical diagnosis and treatment optimization, as well as clinical trial design, accounting for almost 90%of all healthcare data (Beutel et al 2000, Kevin Zhou et al 2020, Shukla-Dave A et al 2019).Medical image analysis techniques aim to enhance the diagnostic performance of visual assessments inmedical imaging, improving the early diagnosis of various diseases and helping to obtain a deeper understanding of physiology and pathology (Beutel et al 2000, Kevin Zhou et al 2020). Medical image computing is an important aspect of computationalmedical imaging, as it aims to directly measure (through end-to-end approaches) or to indirectly guide (via segmentation, synthesis, denoising, reconstruction, etc) the extraction of clinically useful imaging biomarkers (Kevin Zhou et al 2020, Shukla-Dave A et al 2019). Recently, convolutional neural networks (CNN) achieved performance gains across numerous medical image computing tasks (Cheplygina et al 2019,Duncan et al 2019, Yi et al 2019, Chen et al 2021, Kevin Zhou et al 2020). CNNs can efficientlymodel local pixel interactions and be trained on either bigor small-scale data (the latter wasmade possible through data augmentation and other pre-processing techniques) (Yi et al 2019, Kevin Zhou et al 2020). CNNmodels revolutionisedmultiple tasks inmedical image computing, such as image segmentation, registration, reconstruction, denoising, synthesis and pathology detection/evaluation, through extensive analysis ofmedical imaging data (Cheplygina et al 2019,Duncan et al 2019, Yi et al 2019, Haskins et al 2020, Kevin Zhou et al 2020). Depending on themedical image analysis application, the CNNarchitecturesmay vary substantially (Kevin Zhou et al 2020). In this special issue named ‘Focus onmachine learningmodels inmedical imaging’, we aimed to select diversemedical imagingmodalities and downstream clinical applications and encourage the collection of different CNNarchitectures andmethodologies. This special issue collection can be a potentially important and usefulmaterial for future surveys and reviews. It can also be useful towards developing checklists for artificial intelligence criteria inmedical imaging (CLAIM) and a logic (rule-set) for artificial intelligence (AI) algorithmdesign inmedical imaging (Mongan et al 2020), according to themodality and downstream clinical application. The special issue encompasses fifteen peer-reviewed articles covering different imagingmodalities andmodel architectures. Thefirst paper of the special issue byCui et al is a study onmedical image registrationwhich is the task that aims to learn the deformation fieldwhich aligns amovingmedical image to a reference image (from the same or another imagingmodality), to support extracting complementary information and improvemedical image diagnosis (Kunpeng et al 2022). The authors engineered a singular value decomposition (SVD) denoising layer on aU-net architecture. The SVD component helped the registrationmodel to learn low-level image representations and optimally removed noise thus, enhancing registration accuracy. The authors also demonstrated that an exponential linear unit (ELU) activation functionwas robust to noise against other activation function options. Bymaintaining comparable registration times, the authors showed that their registrationmodel performed higher registration accuracy, compared to state-of-the-art (SOTA)methods. The second paper by Zou et al proposed an unsupervisedmotion-compensatedmodel using smoothness regularization onmotion deformationmaps derived fromdifferent time points (images) for the reconstruction of high-resolution free-breathing dynamic lungmagnetic resonance imaging (MRI) (Qing et al 2022).MRI reconstruction throughmachine learning techniques aims to accelerate image acquisitionwithout losing spatiotemporal information, which is critical for clinical imaging. The authors assume the deformationmaps to be the output of a CNN-based generator, whose inputs are time-dependent low-dimensional latent vectors PUBLISHED 16December 2022

Machine Learning in Multimodal Medical Imaging

Multimodal Machine Learning in Image-Based and Clinical Biomedicine: Survey and Prospects

Medical Multimodal Classifiers Under Scarce Data Condition

Beyond Medical Imaging - A Review of Multimodal Deep Learning in Radiology

A review on multimodal machine learning in medical diagnostics

Multimodal Medical Imaging Using Modern Deep Learning Approaches

Medical Multimodal Foundation Models in Clinical Diagnosis and Treatment: Applications, Challenges, and Future Directions

Deep Learning-Based Image Segmentation on Multimodal Medical Imaging

Application of Multimodal Neuroimaging in the Treatment of Neurological Patients

Medical Diagnosis with Large Scale Multimodal Transformers: Leveraging Diverse Data for More Accurate Diagnosis

Medical Imaging Using Machine Learning and Deep Learning Algorithms: A Review

Multi-modal Deep Learning and Its Applications in Ophthalmic Artificial Intelligence

A Survey of Multimodal Machine Learning

Integrating Medical Imaging and Clinical Reports Using Multimodal Deep Learning for Advanced Disease Analysis

A multimodal machine learning model for the stratification of breast cancer risk

Multimodal Machine Learning for Automated ICD Coding.

Machine Learning for Multimodal Electronic Health Records-based Research: Challenges and Perspectives

Focus on Machine Learning Models in Medical Imaging

Machine Learning-Based Multimodel Computing for Medical Imaging for Classification and Detection of Alzheimer Disease

Deep Multimodal Guidance for Medical Image Classification