MRI radiomics based machine learning model of the periaqueductal gray matter in migraine patients.

Background and purpose:

The aim of the study was to investigate the question: Can MRI radiomics analysis of the periaqueductal gray region elucidate the pathophysiological mechanisms underlying various migraine subtypes, and can a machine learning model using these radiomics features accurately differentiate between migraine patients and healthy individuals, as well as between migraine subtypes, including atypical cases with overlapping symptoms?

The study analyzed initial MRI images of individuals taken after their first migraine diagnosis, and additional MRI scans were acquired from healthy subjects. Radiomics modeling was applied to analyze all the MRI images in the periaqueductal gray region. The dataset was randomized, and oversampling was used if there was class imbalance between groups. The optimal algorithm-based feature selection method was employed to select the most important 5-10 features to differentiate between the two groups. The classification performance of AI algorithms was evaluated using receiver operating characteristic analysis to calculate the area under the curve, classification accuracy, sensitivity, and specificity values. Participants were required to have a confirmed diagnosis of either episodic migraine, probable migraine, or chronic migraine. Patients with aura, those who used migraine-preventive medication within the past six months, or had chronic illnesses, psychiatric disorders, cerebrovascular conditions, neoplastic diseases, or other headache types were excluded from the study. Additionally, 102 healthy subjects who met the inclusion and exclusion criteria were included. 

The algorithm-based information gain method for feature reduction had the best performance among all methods, with the first-order, gray-level size zone matrix, and gray-level co-occurrence matrix classes being the dominant feature classes. The machine learning model correctly classified 82.4% of migraine patients from healthy subjects. Within the migraine group, 74.1% of the episodic migraine-probable migraine patients and 90.5% of the chronic migraine patients were accurately classified. No significant difference was found between probable migraine and episodic migraine patients in terms of the periaqueductal gray region radiomics features. The kNN algorithm showed the best performance for classifying episodic migraine-probable migraine subtypes, while the Random Forest algorithm demonstrated the best performance for classifying the migraine group and chronic migraine subtype.

A radiomics-based machine learning model, utilizing standard MR images obtained during the diagnosis and follow-up of migraine patients, shows promise not only in aiding migraine diagnosis and classification for clinical approach, but also in understanding the neurological mechanisms underlying migraines. 

Real-life experiences with galcanezumab and predictors for treatment response in Turkey.

BACKGROUND: The complexity of clinical practice extends far beyond the controlled settings of trials, and there is a need for real-world studies aimed at identifying which patients will respond to anti-CGRP monoclonal antibodies in different countries. This study aimed to investigate the efficacy and safety of galcanezumab in treating migraine in a real-life setting in Turkey, as well as identify predictors of treatment response. METHODS: A total of 476 patients who diagnosed with migraine according to ICHD-3 criteria and treated with galcanezumab by headache specialists were voluntarily participated in this cross-sectional study. Galcanezumab is indicated for the prevention of migraine in adults who have at least 4 monthly migraine days in Turkey. All patients filled out a survey on Google Form that comprised 54 questions, addressing various aspects such as demographics, migraine characteristics, previous use of acute symptomatic medication, failures with preventive drug classes, comorbidities, most bothersome symptoms, as well as the interictal burden of migraine. RESULTS: Among the participants, 89.3% reported that galcanezumab treatment was beneficial for them. A decrease in the frequency (80.0%), severity (85.7%), and acute medication usage for migraine attacks (71.4%) was reported with galcanezumab treatment. An adverse effect related to galcanezumab was reported in 16.3% of cases, but no serious adverse reactions were observed. Remarkably, 14.3% of participants reported no longer experiencing any headaches, and 18.9% did not require any acute treatment while receiving galcanezumab treatment. A logistic regression model showed that male gender, lack of ictal nausea, and previous failure of more than 2 prophylactic agents may predict the non-responders. CONCLUSIONS: The first large series from Turkey showed that galcanezumab treatment is safe and effective in most of the patients diagnosed with migraine by headache experts in the real-life setting. Patients reported a significant decrease in both ictal and interictal burden of migraine and expressed satisfaction with this treatment.

Autoimmune encephalitis and paraneoplastic syndromes in Turkey: a multi-centre study.

BACKGROUND: Autoimmune encephalitis (AIE) and paraneoplastic syndromes (PNS) are both rare groups of neurological diseases that are difficult to diagnose. AIM: We aimed to determine the common and distinct aspects of these two aetiologies of encephalitis as well as the characteristics of our patient group. METHODS: We respectively analysed the records of the patients including symptoms, demographic features, neurological examination, cranial-magnetic-resonance-imaging (MRI), electroencephalography (EEG) findings, cerebrospinal fluid results (CSF) findings. Autoimmune/paraneoplastic autoantibodies in blood and/or CSF were all documented. RESULTS: Forty-six patients fulfilled the diagnostic criteria. Thirty-eight of them were diagnosed with AIE, and 8 of them were diagnosed with PNS. The PNS group had higher nonconvulsive status epilepticus than the AIE (2/8 vs 0/38; p=0.027). PNS patients were diagnosed with a malignancy in their follow-ups more than those in the AIE group [4/38 vs 8/8] (p<0.001). When the symptoms of antibody-positive and negative patients were compared in the AIE group, the rates of consciousness/memory problems (13/15 vs 11/23; p=0.020) and speech impairment (8/15 vs 2/23; p=0.004) were significantly higher in patients without antibodies (n: 15) than in antibody-positive patients (n: 23). In antibody-negative groups, the rates of memory problems in neurological examination (13/15 vs 12/23 p=0.028) and temporal findings on electroencephalography were more prominent than antibody-positive groups (1/23 vs 5/15; p=0.027). The number of patients with cerebellar signs was higher in antibody-positive patients (6/23 vs 0/15; p=0.038). CONCLUSION: Although the positivity of autoantibodies is critical in the diagnosis of AIE and PNS, even minor differences in clinical and laboratory findings of patients are helpful in the diagnosis, especially in the autoantibody-negative patients. Comparing the data with other population studies has shown that several inherited and environmental factors may contribute to the pathophysiology of AIE and PNS, as well as clinical and laboratory differences.

CD4+ and CD25+ T-cell response to short-time interferon-beta therapy on IL10, IL23A and FOXP3 genes in multiple sclerosis patients.

AIM OF THE STUDY: Interferon-beta (IFN-ß), multiple sclerosis (MS) drug for years, does not have therapeutic effects on each patient. Yet, a considerable portion has experienced no therapeutic response to IFN-ß. Therefore, it is necessary to determine disease-specific biomarkers that affect drug response. Here, we aimed to determine the effects of interleukin 10 (IL10) and 23 (IL23A), as well as forkhead box P3 (FOXP3) genes on MS after IFN-ß therapy. MATERIALS AND METHODS: Peripheral blood mononuclear cells (PBMCs) of 42 MS patients were isolated to obtain CD4+ and CD25+ T cells. Both cell types were characterised by flow cytometry. To determine optimum drug concentration of IFN-ß, cytotoxicity assays were assessed on each cell type for 4, 16, 24 and 48 hours respectively. Then, cells were cultured in the presence of 500 IU/mL of IFN-ß. cDNA synthesis was performed after mRNA extraction. RT-PCR was performed to measure gene expressions of IL10, IL23A and FOXP3. Results were evaluated statistically. RESULTS: It was found that the cytotoxic effect of IFN-ß was more efficient as the exposure time was expanded regardless of drug concentration. Moreover, CD25+ T lymphocytes were more resistant to IFN-ß. IL23A was down-regulated, whereas FOXP3 was up-regulated at 48 hours in CD4+ T cells. For CD25+ T cells, the graded increase in FOXP3 was obtained while IL10 expression was gradually decreased throughout the drug intake. CONCLUSION: Although a considerable change in expression was obtained, the long-term IFN-ß effect on both genes and cells should be determined by follow-up at least a year.

Evaluation of language functions in acute cerebellar vascular diseases.

BACKGROUND: The principal features of the cerebellar infarcts are ataxia, failure of coordination, gait instability, and articulation and eye movement disabilities. Language disabilities are also seen with cerebellar lesions, but there are difficulties in diagnosis. This study was planned to evaluate the effects of cerebellar lesions on language functions and the relation between these functions and lesion type, age, and education level. METHODS: A total of 20 patients, 13 male (65%) and 7 female (35%), were included in this study. Twenty control subjects with similar demographic characteristics were also included. The mean age of the patient group was not statistically different. RESULTS: Patients with vermal lesions had significantly higher performance than patients with paravermal lesions when evaluating the understanding of hearing and total aphasia score. Understanding of reading function was significantly better in the patients with small lesions compared with those with large lesions. When the aphasia parameters were compared between the patient and control groups, significant differences were found for all parameters, which reflected the language abilities understanding, naming, true-wrong questions, complex questions, comparing, repeating, and total aphasia score. Reading and writing functions were also significantly different between the patients and the control subjects. CONCLUSION: We found that the cerebellum contributes in several language parameters. These functions show difference according to localization and lesion volume of cerebellar disease. Aphasia should be explored as an important parameter when evaluating the loss of function in patients with cerebellar lesions.