Tumefactive Demyelination Lesions: Report on three cases.

PURPOSE: Tumefactive demyelination (TD) lesion and its subtype Balo's concentric sclerosis (BCS), are rare manifestations of central nervous system demyelinating disease. Because of its rarity, physicians might hesitate in reaching a diagnosis or initiating steroid pulse therapy. This study aims at pinpointing the key neuroimaging features to distinguish TD lesions from surgical conditions, and illustrating the clinical outcomes of patients with TD lesions. CASE REPORT: Two of the three patients had solitary TD lesions, one 47-year-old man presenting with newly onset seizure and another 54-year-old women suffering from progressive hemiparesis. The male patient underwent craniotomy for mass excision without further steroid therapy, while the female patient received methylprednisolone pulse therapy only. Both patients remained free of clinical and radiological relapses over the past 6-7 years, leading to the diagnosis of clinically isolated syndrome. The third case is a 30-year-old woman with subacute onset of dysarthria and hemiparesis. She had two BCS lesions along with other demyelinating lesions in the juxtacortical and periventricular regions, cerebellar peduncles, and spinal cord, fulfilling dissemination in time and space. Her neurological deficits resolved after pulse therapy, and she received long-term disease modifying therapy for multiple sclerosis. CONCLUSION: This study underscores the diverse neuroimaging and clinical presentations of patients with TD lesions, and emphasizes the importance of clinical vigilance regarding this rare condition.

Development of Machine Learning Models for Prediction of Osteoporosis from Clinical Health Examination Data.

Osteoporosis is treatable but often overlooked in clinical practice. We aimed to construct prediction models with machine learning algorithms to serve as screening tools for osteoporosis in adults over fifty years old. Additionally, we also compared the performance of newly developed models with traditional prediction models. Data were acquired from community-dwelling participants enrolled in health checkup programs at a medical center in Taiwan. A total of 3053 men and 2929 women were included. Models were constructed for men and women separately with artificial neural network (ANN), support vector machine (SVM), random forest (RF), k-nearest neighbor (KNN), and logistic regression (LoR) to predict the presence of osteoporosis. Area under receiver operating characteristic curve (AUROC) was used to compare the performance of the models. We achieved AUROC of 0.837, 0.840, 0.843, 0.821, 0.827 in men, and 0.781, 0.807, 0.811, 0.767, 0.772 in women, for ANN, SVM, RF, KNN, and LoR models, respectively. The ANN, SVM, RF, and LoR models in men, and the ANN, SVM, and RF models in women performed significantly better than the traditional Osteoporosis Self-Assessment Tool for Asians (OSTA) model. We have demonstrated that machine learning algorithms improve the performance of screening for osteoporosis. By incorporating the models in clinical practice, patients could potentially benefit from earlier diagnosis and treatment of osteoporosis.