Correctly identifying gait phases is a prerequisite to achieve a spatial/temporal characterization of muscular recruitment during walking. Recent approaches have addressed this issue by applying machine learning techniques to treadmill-walking data. We propose a deep learning approach for surface electromyographic (sEMG)-based classification of stance/swing phases and prediction of the foot–floor-contact signal in more natural walking conditions (similar to everyday walking ones), overcoming constraints of a controlled environment, such as treadmill walking. To this aim, sEMG signals were acquired from eight lower-limb muscles in about 10.000 strides from 23 healthy adults during level ground walking, following an eight-shaped path including natural deceleration, reversing, curve, and acceleration. By means of an extensive evaluation, we show that using a multi layer perceptron to learn hidden features provides state of the art performances while avoiding features engineering. Results, indeed, showed an average classification accuracy of $94.9$ for learned subjects and $93.4$ for unlearned ones, while mean absolute difference ( $\pm SD$ ) between phase transitions timing predictions and footswitch data was $21.6$ ms and $38.1$ ms for heel-strike and toe off, respectively. The suitable performance achieved by the proposed method suggests that it could be successfully used to automatically classify gait phases and predict foot–floor-contact signal from sEMG signals during level ground walking.

Background and Objective: In 2004, the World Health Organization defined atypical pituitary adenomas as those with a Ki-67 expression >3%, an excessive p53 expression and increased mitotic activity. As the usefulness of this classification is controversial, we reviewed typical and atypical pituitary adenomas to compare the clinical and prognostic features. Patients and Methods: We retrospectively reviewed 343 consecutive pituitary adenomas. Atypical pituitary adenomas represented 18.7% of cases. All patients were operated on at the Department of Neurosurgery of our institution and were followed up at the Hypothalamic-Pituitary Disease Unit of the same institution. The median follow-up time was 75 months (range 7-345). Results: Younger age at diagnosis as well as immunohistochemical positivity for adrenocorticotropic hormone and prolactin correlated with a higher risk of atypical pituitary adenomas, whereas typical and atypical pituitary adenomas did not differ with regard to gender, tumor size, recurrence risk and disease-free survival time (DFST). Among the 219 patients who underwent radical surgery, a Ki-67 expression ≥1.5% was associated with a higher risk of recurrence and a worse DFST, even after correction for age at diagnosis, gender, immunohistochemical classification, tumor size, invasiveness and Knosp classification [p = 0.01; hazard ratio (HR) 2.572; 95% confidence interval (CI) 1.251-5.285). Pituitary adenomas with a Ki-67 expression ≥1.5% showed a worse DFST as compared to pituitary adenomas with a Ki-67 expression <1.5% (HR 2.166; 95% CI 1.154-4.064). Conclusion: In this series, atypical and typical pituitary adenomas did not differ with regard to recurrence and DFST. Pituitary adenomas with a Ki-67 expression ≥1.5% showed a higher recurrence risk and a worse DFST as compared to those with a Ki-67 expression <1.5%. We suggest that a Ki-67 expression ≥1.5% may be useful as a prognostic marker, though this will need to be confirmed by prospective, multicenter data.