METHODS: All relevant studies were identified through keyword searches in electronic databases from inception until September 2020. The searched publications were reviewed, categorised and analysed based on their respective methodology.

RESULTS: Hundred and one publications were identified which utilised existing MC-based applications/programs or customised MC simulations. Two outstanding challenges were identified that contribute to uncertainties in the virtual simulation reconstruction. The first challenge involves the use of anatomical models to represent individuals. Currently, phantom libraries best balance the needs of clinical practicality with those of specificity. However, mismatches of anatomical variations including body size and organ shape can create significant discrepancies in dose estimations. The second challenge is that the exact positioning of the patient relative to the beam is generally unknown. Most dose prediction models assume the patient is located centrally on the examination couch, which can lead to significant errors.

METHODS: A total of 1447 ultrasound images, including 767 benign masses and 680 malignant masses were acquired from a tertiary hospital. A semi-supervised GAN model was developed to augment the breast ultrasound images. The synthesized images were subsequently used to classify breast masses using a convolutional neural network (CNN). The model was validated using a 5-fold cross-validation method.

RESULTS: The proposed GAN architecture generated high-quality breast ultrasound images, verified by two experienced radiologists. The improved performance of semi-supervised learning increased the quality of the synthetic data produced in comparison to the baseline method. We achieved more accurate breast mass classification results (accuracy 90.41%, sensitivity 87.94%, specificity 85.86%) with our synthetic data augmentation compared to other state-of-the-art methods.

METHODS: A phantom study was performed to investigate the correlation of (1)H MRS-visible lipids with the signal loss ratio (SLR) obtained using IOP imaging. A cross-sectional study approved by the institutional review board was carried out in 22 patients with different glioma grades. The patients underwent scanning using IOP imaging and single-voxel spectroscopy (SVS) using 3T MRI. The brain spectra acquisitions from solid and cystic components were obtained and correlated with the SLR for different grades.

Methods: Forty histologically proven glioma patients underwent a standard MRI tumour protocol with the addition of IOP sequence. The regions of tumour (solid enhancing, solid non-enhancing, and cystic regions) were delineated using snake model (ITK-SNAP) with reference to structural and diffusion MRI images. The lipid distribution map was constructed based on signal loss ratio (SLR) obtained from the IOP imaging. The mean SLR values of the regions were computed and compared across the different glioma grades.

METHODS: We scrutinized the routine radiological exposure parameters during 58 clinical neuro-interventional procedures such as, exposure direction, magnification, frame rate, and distance between image receptor to patient's body and evaluate their effects on patient's dose using an anthropomorphic phantom. Radiation dose received by the occipital region, ears and eyes of the phantom were measured using MOSkin detectors.

RESULTS: DSA imaging technique is a major contributor to patient's dose (80.9%) even though they are used sparingly (5.3% of total frame number). The occipital region of the brain received high dose largely from the frontal tube constantly placed under couch (73.7% of the total KAP). When rotating the frontal tube away from under the couch, the radiation dose to the occipital reduced by 40%. The use of magnification modes could increase radiation dose by 94%. Changing the image receptor to the phantom surface distance from 10 to 40cm doubled the radiation dose received by the patient's skin at the occipital region.

METHODS: The study comprised 106 chronic kidney disease (CKD) patients and 203 control subjects. Conventional ultrasound was performed to measure the kidney length and cortical thickness. SWE imaging was performed to measure renal parenchymal stiffness. Diagnostic performance of SWE and conventional ultrasound were correlated with serum creatinine, urea levels and eGFR.

RESULTS: Pearson's correlation coefficient revealed a negative correlation between YM measurements and eGFR (r = -0.576, p < 0.0001). Positive correlations between YM measurements and age (r = 0.321, p < 0.05), serum creatinine (r = 0.375, p < 0.0001) and urea (r = 0.287, p < 0.0001) were also observed. The area under the receiver operating characteristic curve for SWE (0.87) was superior to conventional ultrasound alone (0.35-0.37). The cut-off value of less or equal to 4.31 kPa suggested a non-diseased kidney (80.3% sensitivity, 79.5% specificity).

METHODS: 41 medical personnel performing 79 procedures were monitored for their eye lens exposure using the NanoDot™ optically-stimulated luminescence dosimeters (OSLD) taped to the outer canthus of their eyes. The air-kerma area product (KAP), fluoroscopy time (FT) and number of procedure runs were also recorded.

RESULTS: KAP, FT and number of runs were strongly correlated. However, only weak to moderate correlations were observed between these parameters with the measured eye lens doses. The average median equivalent eye lens dose was 0.052 mSv (ranging from 0.0155 to 0.672 mSv). The eye lens doses of primary operators were found to be significantly higher than their assistants due to the closer proximity to the patient and X-ray tube. The left eye lens of the operators received the highest amount of radiation due to their habitual positioning towards the radiation source.

METHODS: We performed a literature search of published articles on the application of MRI phenotypic features in invasive breast cancer molecular subtype classifications by radiologists' interpretation on Medline Complete, Pubmed, and Google scholar from 1st January 2000 to 31st March 2021. Of the 1453 literature identified, 42 fulfilled the inclusion criteria.

RESULTS: All studies were case-controlled, retrospective study and research-based. The majority of the studies assessed the MRI features using American College of Radiology- Breast Imaging Reporting and Data System (ACR-BIRADS) classification and using dynamic contrast-enhanced (DCE) kinetic features, Apparent Diffusion Coefficient (ADC) values, and T2 sequence. Most studies divided invasive breast cancer into 4 main subtypes, luminal A, luminal B, HER2, and triple-negative (TN) cancers, and used 2 readers. We present a summary of the radiologists' extracted breast MRI phenotypical features and their correlating breast cancer subtypes classifications. The characteristic features are morphology, enhancement kinetics, and T2 signal intensity. We found that the TN subtype has the most distinctive MRI features compared to the other subtypes and luminal A and B have many similar features.