RESUMEN
The detection of contrast-enhancing lesions (CELs) is fundamental for the diagnosis and monitoring of patients with multiple sclerosis (MS). This task is time-consuming and suffers from high intra- and inter-rater variability in clinical practice. However, only a few studies proposed automatic approaches for CEL detection. This study aimed to develop a deep learning model that automatically detects and segments CELs in clinical Magnetic Resonance Imaging (MRI) scans. A 3D UNet-based network was trained with clinical MRI from the Swiss Multiple Sclerosis Cohort. The dataset comprised 372 scans from 280 MS patients: 162 showed at least one CEL, while 118 showed no CELs. The input dataset consisted of T1-weighted before and after gadolinium injection, and FLuid Attenuated Inversion Recovery images. The sampling strategy was based on a white matter lesion mask to confirm the existence of real contrast-enhancing lesions. To overcome the dataset imbalance, a weighted loss function was implemented. The Dice Score Coefficient and True Positive and False Positive Rates were 0.76, 0.93, and 0.02, respectively. Based on these results, the model developed in this study might well be considered for clinical decision support.
RESUMEN
INTRODUCTION: Several disease-modifying therapies (DMTs) show efficacy in relapsing-remitting multiple sclerosis (RRMS) and clinically isolated syndrome (CIS). However, there is still a relevant proportion of patients who remain untreated. We provide real-world data on untreated and treated patients and we report whether and how the introduction of oral DMTs changed the treatment decision. Furthermore, we discuss possible reasons for not receiving DMTs. METHODS: We conducted a retrospective cross-sectional study and analysed demographic and clinical data of patients with RRMS and CIS at our MS center. Comparison was made between untreated and treated patients in 2010 (before the introduction of oral DMTs) and in 2014 (after the introduction of oral DMTs). Furthermore, we analysed reasons for the decision against DMTs in patients who never received DMTs and patients who discontinued DMTs. RESULTS: We analysed datasets of 344 MS patients in 2010 and 253 in 2014. There were more untreated patients in CIS than in RRMS. In RRMS, the proportion of untreated patients decreased significantly between 2010 and 2014 from 23.6% to 11.1%, while the use of oral medications increased significantly from <1% to more than 50% in 2014. In CIS, there was no significant change in untreated patients between 2010 and 2014 (61.1% in 2010 to 52.6% in 2014). Untreated patients with RRMS were significantly older and had lower ARR than treated patients. Patients who never received DMT had lower EDSS compared to patients that had been treated before. The main reasons for the decision against DMT were "belief in a benign course" and "fear of adverse effects". Treatment discontinuation was caused mainly by the adverse effects. DISCUSSION: In our data a relevant proportion of patients with RRMS and CIS did not receive any DMT. We hypothesize that in patients with RRMS the introduction of oral DMTs translated to a higher rate of treatment, whereas in CIS there no change was observed. This could be due to limited therapeutic options in CIS. There is more information needed regarding the treatment recommendation for older patients and patients with mild course of the disease.
Asunto(s)
Esclerosis Múltiple Recurrente-Remitente , Esclerosis Múltiple , Humanos , Esclerosis Múltiple Recurrente-Remitente/tratamiento farmacológico , Estudios Transversales , Estudios Retrospectivos , SuizaRESUMEN
In this case study with video and neurophysiology, we describe a rare case of hemimyorhythmia occurring 4 months after a stroke with bilateral affection of the thalamus and right superior cerebellar peduncle (Guillain-Mollaret-triangle). This case and especially the video with the clinical and EMG presentation of a synchronous rhythmic pattern at 3,1 Hz makes an important educational contribution to the recognition of myorhythmia and discussed differential diagnoses.