Dir Journal Diagnostic and Interventional Radiology

Diagnostic and Interventional Radiology

We sought to present our preliminary experience on the effectiveness and safety of magnetic resonance imaging (MRI)-guided, high-intensity focused ultrasound (HIFU) therapy using a volumetric ablation technique in the treatment of Association of Asian Nations (ASEAN) patients with symptomatic uterine leiomyomas. This study included 33 women who underwent HIFU treatment. Tissue characteristics of leiomyomas were assessed based on T2- and T1-weighted MRI. The immediate nonperfused volume (NPV) ratio and the treatment effectiveness of MRI-guided HIFU on the basis of the degrees of volume reduction and improvement in transformed symptom severity score (SSS) were assessed. The median immediate NPV ratio was 89.8%. Additionally, the median acoustic sonication power and HIFU treatment durations were 150 W and 125 min, respectively. At six-month follow-up, the median leiomyoma volume had decreased from 139 mL at baseline to 84 mL and the median transformed SSS had decreased from 56.2 at baseline to 18.8. No major adverse events were observed. The preliminary results demonstrated that volumetric MRI-guided HIFU therapy for the treatment of symptomatic leiomyomas in ASEAN patients appears to be clinically acceptable with regard to treatment effectiveness and safety.

Background: Hospital-based stroke rehabilitation for stroke survivors in developing countries may be limited by staffing ratios and length of stay that could hamper recovery potential. Thus, a home-based, gamified rehabilitation system (i.e., IntelliRehab) was tested for its ability to increase cerebral blood flow (CBF), and the secondary impact of changes on the upper limb motor function and functional outcomes. Objective: To explore the effect of IntelliRehab on CBF in chronic stroke patients and its correlation with the upper limb motor function. Methods: Two-dimensional pulsed Arterial Spin Labelling (2D-pASL) was used to obtain CBF images of stable, chronic stroke subjects (n = 8) over 3-months intervention period. CBF alterations were mapped, and the detected differences were marked as regions of interest. Motor functions represented by Fugl-Meyer Upper Extremity Assessment (FMA) and Stroke Impact Scale (SIS) were used to assess the primary and secondary outcomes, respectively. Results: Regional CBF were significantly increased in right inferior temporal gyrus and left superior temporal white matter after 1-month (p = 0.044) and 3-months (p = 0.01) of rehabilitation, respectively. However, regional CBF in left middle fronto-orbital gyrus significantly declined after 1-month of rehabilitation (p = 0.012). Moreover, SIS-Q7 and FMA scores significantly increased after 1-month and 3-months of rehabilitation. There were no significant correlations, however, between CBF changes and upper limb motor function. Conclusions: Participants demonstrated improved motor functions, supporting the benefit of using IntelliRehab as a tool for home-based rehabilitation. However, within-participant improvements may have limited potential that suggests the need for a timely administration of IntelliRehab to get the maximum capacity of improvement.

Machine learning (ML) models have become capable of making critical decisions on our behalf. Nevertheless, due to complexity of these models, interpreting their decisions can be challenging, and humans cannot always control them. This paper provides explanations of decisions made by ML models in diagnosing four types of posterior fossa tumors: medulloblastoma, ependymoma, pilocytic astrocytoma, and brainstem glioma. The proposed methodology involves data analysis using kernel density estimations with Gaussian distributions to examine individual MRI features, conducting an analysis on the relationships between these features, and performing a comprehensive analysis of ML model behavior. This approach offers a simple yet informative and reliable means of identifying and validating distinguishable MRI features for the diagnosis of pediatric brain tumors. By presenting a comprehensive analysis of the responses of the four pediatric tumor types to each other and to ML models in a single source, this study aims to bridge the knowledge gap in the existing literature concerning the relationship between ML and medical outcomes. The results highlight that employing a simplistic approach in the absence of very large datasets leads to significantly more pronounced and explainable outcomes, as expected. Additionally, the study also demonstrates that the pre-analysis results consistently align with the outputs of the ML models and the clinical findings reported in the existing literature.