Recent trends in AI applications for pelvic MRI: a comprehensive review.

Magnetic resonance imaging (MRI) is an essential tool for evaluating pelvic disorders affecting the prostate, bladder, uterus, ovaries, and/or rectum. Since the diagnostic pathway of pelvic MRI can involve various complex procedures depending on the affected organ, the Reporting and Data System (RADS) is used to standardize image acquisition and interpretation. Artificial intelligence (AI), which encompasses machine learning and deep learning algorithms, has been integrated into both pelvic MRI and the RADS, particularly for prostate MRI. This review outlines recent developments in the use of AI in various stages of the pelvic MRI diagnostic pathway, including image acquisition, image reconstruction, organ and lesion segmentation, lesion detection and classification, and risk stratification, with special emphasis on recent trends in multi-center studies, which can help to improve the generalizability of AI.

Frontoparietal-Striatal Network and Nucleus Basalis Modulation in Patients With Parkinson Disease and Gait Disturbance.

BACKGROUND AND OBJECTIVES: Neural computations underlying gait disorders in Parkinson disease (PD) are multifactorial and involve impaired expression of stereotactic locomotor patterns and compensatory recruitment of cognitive functions. This study aimed to clarify the network mechanisms of cognitive contribution to gait control and its breakdown in patients with PD. METHODS: Patients with PD were instructed to walk at a comfortable pace on a mat with pressure sensors. The characterization of cognitive-motor interplay was enhanced by using a gait with a secondary cognitive task (dual-task condition) and a gait without additional tasks (single-task condition). Participants were scanned using 3-T MRI and 123I-ioflupane SPECT. RESULTS: According to gait characteristics, cluster analysis assisted by a nonlinear dimensionality reduction technique, t-distributed stochastic neighbor embedding, categorized 56 patients with PD into 3 subpopulations. The preserved gait (PG) subgroup (n = 23) showed preserved speed and variability during gait, both with and without additional cognitive load. Compared with the PG subgroup, the mildly impaired gait (MIG) subgroup (n = 16) demonstrated deteriorated gait variability with additional cognitive load and impaired speed and gait variability without additional cognitive load. The severely impaired gait (SIG) subgroup (n = 17) revealed the slowest speed and highest gait variability. In addition, group differences were found in attention/working memory and executive function domains, with the lowest performance in the SIG subgroup than in the PG and MIG subgroups. Using resting-state functional MRI, the SIG subgroup demonstrated lower functional connectivity of the left and right frontoparietal network (FPN) with the caudate than the PG subgroup did (left FPN, d = 1.21, p < 0.001; right FPN, d = 1.05, p = 0.004). Cortical thickness in the FPN and 123I-ioflupane uptake in the striatum did not differ among the 3 subgroups. By contrast, the severity of Ch4 density loss was significantly correlated with the level of functional connectivity degradation of the FPN and caudate (left FPN-caudate, r = 0.27, p = 0.04). DISCUSSION: These findings suggest that the functional connectivity of the FPN with the caudate, as mediated by the cholinergic Ch4 projection system, underlies the compensatory recruitment of attention and executive function for damaged automaticity in gait in patients with PD.

Climate change and artificial intelligence in healthcare: Review and recommendations towards a sustainable future.

The rapid advancement of artificial intelligence (AI) in healthcare has revolutionized the industry, offering significant improvements in diagnostic accuracy, efficiency, and patient outcomes. However, the increasing adoption of AI systems also raises concerns about their environmental impact, particularly in the context of climate change. This review explores the intersection of climate change and AI in healthcare, examining the challenges posed by the energy consumption and carbon footprint of AI systems, as well as the potential solutions to mitigate their environmental impact. The review highlights the energy-intensive nature of AI model training and deployment, the contribution of data centers to greenhouse gas emissions, and the generation of electronic waste. To address these challenges, the development of energy-efficient AI models, the adoption of green computing practices, and the integration of renewable energy sources are discussed as potential solutions. The review also emphasizes the role of AI in optimizing healthcare workflows, reducing resource waste, and facilitating sustainable practices such as telemedicine. Furthermore, the importance of policy and governance frameworks, global initiatives, and collaborative efforts in promoting sustainable AI practices in healthcare is explored. The review concludes by outlining best practices for sustainable AI deployment, including eco-design, lifecycle assessment, responsible data management, and continuous monitoring and improvement. As the healthcare industry continues to embrace AI technologies, prioritizing sustainability and environmental responsibility is crucial to ensure that the benefits of AI are realized while actively contributing to the preservation of our planet.

Segmentation and Volume Estimation of the Habenula Using Deep Learning in Patients With Depression.

Background: The habenula is involved in the pathophysiology of depression. However, its small structure limits the accuracy of segmentation methods, and the findings regarding its volume have been inconsistent. This study aimed to create a highly accurate habenula segmentation model using deep learning, test its generalizability to clinical magnetic resonance imaging, and examine differences between healthy participants and patients with depression. Methods: This multicenter study included 382 participants (patients with depression: N = 234, women 47.0%; healthy participants: N = 148, women 37.8%). A 3-dimensional residual U-Net was used to create a habenula segmentation model on 3T magnetic resonance images. The reproducibility and generalizability of the predictive model were tested on various validation cohorts. Thereafter, differences between the habenula volume of healthy participants and that of patients with depression were examined. Results: A Dice coefficient of 86.6% was achieved in the derivation cohort. The test-retest dataset showed a mean absolute percentage error of 6.66, indicating sufficiently high reproducibility. A Dice coefficient of >80% was achieved for datasets with different imaging conditions, such as magnetic field strengths, spatial resolutions, and imaging sequences, by adjusting the threshold. A significant negative correlation with age was observed in the general population, and this correlation was more pronounced in patients with depression (p < 10-7, r = -0.59). Habenula volume decreased with depression severity in women even when the effects of age and scanner were excluded (p = .019, η2 = 0.099). Conclusions: Habenula volume could be a pathophysiologically relevant factor and diagnostic and therapeutic marker for depression, particularly in women.

Accurate segmentation of the habenula, a brain region implicated in depression, is challenging. In this study, we developed an automated human habenula segmentation model using deep learning techniques. The model was confirmed to be reproducible and generalizable at various spatial resolutions. Application of this model to a multicenter dataset confirmed that habenula volume decreased with age in healthy volunteers, an association that was more pronounced in individuals with depression. In addition, habenula volume decreased with the severity of depression in women. This novel model for habenula segmentation enables further study of the role of the habenula in depression.

Quantitative susceptibility mapping and a nonlinearly transformed atlas for targeting the ventral intermediate nucleus of the thalamus in a patient with tremor and thalamic hypertrophy: illustrative case.

BACKGROUND: The ventral intermediate nucleus (Vim) of the thalamus is a surgical target for treating various types of tremor. Because it is difficult to visualize the Vim using standard magnetic resonance imaging, the structure is usually targeted based on the anterior and posterior commissures. This standard targeting method is practical in most patients but not in those with thalamic asymmetry. The authors examined the usefulness of quantitative susceptibility mapping (QSM) and transformed Vim atlas images to estimate the Vim localization in a patient with tremor and significant thalamic hypertrophy. OBSERVATIONS: A 51-year-old right-handed female had experienced a predominant left-hand action tremor for 6 years. Magnetic resonance imaging showed significant hypertrophy of the right thalamus and caudal shift of the thalamic ventral border. The authors referred to the QSM images to localize the decreased susceptibility area within the lateral ventral thalamic nuclei to target the Vim. In addition, the nonlinearly transformed Vim atlas images complemented the imaging-based targeting. The radiofrequency thalamotomy at the modified Vim target relieved the tremor completely. LESSONS: A combination of QSM and nonlinear transformation of the thalamic atlas can be helpful in the targeting method of the Vim for tremor patients with thalamic asymmetry.

Severe Bronchopulmonary Dysplasia Adversely Affects Brain Growth in Preterm Infants.

INTRODUCTION: Bronchopulmonary dysplasia (BPD) is associated with neurodevelopmental outcomes of preterm infants, but its effect on brain growth in preterm infants after the neonatal period is unknown. This study aimed to evaluate the effect of severe BPD on brain growth of preterm infants from term to 18 months of corrected age (CA). METHODS: Sixty-three preterm infants (42 with severe BPD and 21 without severe BPD) who underwent magnetic resonance imaging at term equivalent age (TEA) and 18 months of CA were studied by using the Infant Brain Extraction and Analysis Toolbox (iBEAT). We measured segmented brain volumes and compared brain volume and brain growth velocity between the severe BPD group and the non-severe BPD group. RESULTS: There was no significant difference in brain volumes at TEA between the groups. However, the brain volumes of the total brain and cerebral white matter in the severe BPD group were significantly smaller than those in the non-severe BPD group at 18 months of CA. The brain growth velocities from TEA to 18 months of CA in the total brain, cerebral cortex, and cerebral white matter in the severe BPD group were lower than those in the non-severe BPD group. CONCLUSION: Brain growth in preterm infants with severe BPD from TEA age to 18 months of CA is less than that in preterm infants without severe BPD.

Clinical and imaging characteristics of supratentorial glioma with IDH2 mutation.

PURPOSE: The rarity of IDH2 mutations in supratentorial gliomas has led to gaps in understanding their radiological characteristics, potentially resulting in misdiagnosis based solely on negative IDH1 immunohistochemical staining. We aimed to investigate the clinical and imaging characteristics of IDH2-mutant gliomas. METHODS: We analyzed imaging data from adult patients with pathologically confirmed diffuse lower-grade gliomas and known IDH1/2 alteration and 1p/19q codeletion statuses obtained from the records of our institute (January 2011 to August 2022, Cohort 1) and The Cancer Imaging Archive (TCIA, Cohort 2). Two radiologists evaluated clinical information and radiological findings using standardized methods. Furthermore, we compared the data for IDH2-mutant and IDH-wildtype gliomas. Multivariate logistic regression was used to identify the predictors of IDH2 mutation status, and receiver operating characteristic curve analysis was employed to assess the predictive performance of the model. RESULTS: Of the 20 IDH2-mutant supratentorial gliomas, 95% were in the frontal lobes, with 75% classified as oligodendrogliomas. Age and the T2-FLAIR discordance were independent predictors of IDH2 mutations. Receiver operating characteristic curve analysis for the model using age and T2-FLAIR discordance demonstrated a strong potential for discriminating between IDH2-mutant and IDH-wildtype gliomas, with an area under the curve of 0.96 (95% CI, 0.91-0.98, P = .02). CONCLUSION: A high frequency of oligodendrogliomas with 1p/19q codeletion was observed in IDH2-mutated gliomas. Younger age and the presence of the T2-FLAIR discordance were associated with IDH2 mutations and these findings may help with precise diagnoses and treatment decisions in clinical practice.

Thin-slice reverse encoding distortion correction DWI facilitates visualization of non-functioning pituitary neuroendocrine tumor (PitNET)/pituitary adenoma and surrounding normal structures.

BACKGROUND: To evaluate the clinical usefulness of thin-slice echo-planar imaging (EPI)-based diffusion-weighted imaging (DWI) with an on-console distortion correction technique, termed reverse encoding distortion correction DWI (RDC-DWI), in patients with non-functioning pituitary neuroendocrine tumor (PitNET)/pituitary adenoma. METHODS: Patients with non-functioning PitNET/pituitary adenoma who underwent 3-T RDC-DWI between December 2021 and September 2022 were retrospectively enrolled. Image quality was compared among RDC-DWI, DWI with correction for distortion induced by B0 inhomogeneity alone (B0-corrected-DWI), and original EPI-based DWI with anterior-posterior phase-encoding direction (AP-DWI). Susceptibility artifact, anatomical visualization of cranial nerves, overall tumor visualization, and visualization of cavernous sinus invasion were assessed qualitatively. Quantitative assessment of geometric distortion was performed by evaluation of anterior and posterior displacement between each DWI and the corresponding three-dimensional T2-weighted imaging. Signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and apparent diffusion coefficient values were measured. RESULTS: Sixty-four patients (age 70.8 ± 9.9 years [mean ± standard deviation]; 33 females) with non-functioning PitNET/pituitary adenoma were evaluated. In terms of susceptibility artifacts in the frontal and temporal lobes, visualization of left trigeminal nerve, overall tumor visualization, and anterior displacement, RDC-DWI performed the best and B0-corrected-DWI performed better than AP-DWI. The right oculomotor and right trigeminal nerves were better visualized by RDC-DWI than by B0-corrected-DWI and AP-DWI. Visualization of cavernous sinus invasion and posterior displacement were better by RDC-DWI and B0-corrected-DWI than by AP-DWI. SNR and CNR were the highest for RDC-DWI. CONCLUSIONS: RDC-DWI achieved excellent image quality regarding susceptibility artifact, geometric distortion, and tumor visualization in patients with non-functioning PitNET/pituitary adenoma. RELEVANCE STATEMENT: RDC-DWI facilitates excellent visualization of the pituitary region and surrounding normal structures, and its on-console distortion correction technique is convenient. RDC-DWI can clearly depict cavernous sinus invasion of PitNET/pituitary adenoma even without contrast medium. KEY POINTS: • RDC-DWI is an EPI-based DWI technique with a novel on-console distortion correction technique. • RDC-DWI corrects distortion due to B0 field inhomogeneity and eddy current. • We evaluated the usefulness of thin-slice RDC-DWI in non-functioning PitNET/pituitary adenoma. • RDC-DWI exhibited excellent visualization in the pituitary region and surrounding structures. • In addition, the on-console distortion correction of RDC-DWI is clinically convenient.

Specific consistency score for rational selection of epilepsy resection surgery candidates.

OBJECTIVE: Degree of indication for epilepsy surgery is determined by taking multiple factors into account. This study aimed to investigate the usefulness of the Specific Consistency Score (SCS), a proposed score for focal epilepsy to rate the indication for epilepsy focal resection. METHODS: This retrospective cohort study included patients considered for resective epilepsy surgery in Kyoto University Hospital from 2011 to 2022. Plausible epileptic focus was tentatively defined. Cardinal findings were scored based on specificity and consistency with the estimated laterality and lobe. The total points represented SCS. The association between SCS and the following clinical parameters was assessed by univariate and multivariate analysis: (1) probability of undergoing resective epilepsy surgery, (2) good postoperative seizure outcome (Engel I and II or Engel I only), and (3) lobar concordance between the noninvasively estimated focus and intracranial electroencephalographic (EEG) recordings. RESULTS: A total of 131 patients were evaluated. Univariate analysis revealed higher SCS in the (1) epilepsy surgery group (8.4 [95% confidence interval (CI) = 7.8-8.9] vs. 4.9 [95% CI = 4.3-5.5] points; p < .001), (2) good postoperative seizure outcome group (Engel I and II; 8.7 [95% CI = 8.2-9.3] vs. 6.4 [95% CI = 4.5-8.3] points; p = .008), and (3) patients whose focus defined by intracranial EEG matched the noninvasively estimated focus (8.3 [95% CI = 7.3-9.2] vs. 5.4 [95% CI = 3.5-7.3] points; p = .004). Multivariate analysis revealed areas under the curve of .843, .825, and .881 for Parameters 1, 2, and 3, respectively. SIGNIFICANCE: SCS provides a reliable index of good indication for resective epilepsy surgery and can be easily available in many institutions not necessarily specializing in epilepsy.

The impact of large language models on radiology: a guide for radiologists on the latest innovations in AI.

The advent of Deep Learning (DL) has significantly propelled the field of diagnostic radiology forward by enhancing image analysis and interpretation. The introduction of the Transformer architecture, followed by the development of Large Language Models (LLMs), has further revolutionized this domain. LLMs now possess the potential to automate and refine the radiology workflow, extending from report generation to assistance in diagnostics and patient care. The integration of multimodal technology with LLMs could potentially leapfrog these applications to unprecedented levels.However, LLMs come with unresolved challenges such as information hallucinations and biases, which can affect clinical reliability. Despite these issues, the legislative and guideline frameworks have yet to catch up with technological advancements. Radiologists must acquire a thorough understanding of these technologies to leverage LLMs' potential to the fullest while maintaining medical safety and ethics. This review aims to aid in that endeavor.

Ruptured thrombosed vertebral artery dissecting aneurysm treated with staged flow diverter after prediction of the rupture point by vessel wall MRI.

The safety and feasibility of using staged flow diverter (FD) for ruptured cerebral aneurysms, in which coil embolization is performed in the acute phase and FD is deployed in the subacute phase, has recently been reported. This strategy requires assuming the rupture point and performing coil embolization. Although vessel wall magnetic resonance imaging (VW-MRI) has been reported to be useful in predicting the rupture point of aneurysms, its use with staged FD has not yet been reported. We report the first case of staged FD with preoperative contrast-enhanced VW-MRI to predict the rupture point for partially thrombosed vertebral artery dissecting large aneurysm involving posterior inferior cerebellar artery (PICA) origin. This approach achieved a very good outcome, not only completely occluding the aneurysm, but also reconstructing the parent artery while maintaining the patency of the PICA.