Structure of the set of feasible neural commands for complex motor tasks.

The brain must select its control strategies among an infinite set of possibilities; researchers believe that it must be solving an optimization problem. While this set of feasible solutions is infinite and lies in high dimensions, it is bounded by kinematic, neuromuscular, and anatomical constraints, within which the brain must select optimal solutions. That is, the set of feasible activations is well structured. However, to date there is no method to describe and quantify the structure of these high-dimensional solution spaces. Bounding boxes or dimensionality reduction algorithms do not capture their detailed structure. We present a novel approach based on the well-known Hit-and-Run algorithm in computational geometry to extract the structure of the feasible activations capable of producing 50% of maximal fingertip force in a specific direction. We use a realistic model of a human index finger with 7 muscles, and 4 DOFs. For a given static force vector at the endpoint, the feasible activation space is a 3D convex polytope, embedded in the 7D unit cube. It is known that explicitly computing the volume of this polytope can become too computationally complex in many instances. However, our algorithm was able to sample 1,000,000 uniform at random points from the feasible activation space. The computed distribution of activation across muscles sheds light onto the structure of these solution spaces-rather than simply exploring their maximal and minimal values. Although this paper presents a 7 dimensional case of the index finger, our methods extend to systems with at least 40 muscles. This will allow our motor control community to understand the distributions of feasible muscle activations, providing important contextual information into learning, optimization and adaptation of motor patterns in future research.

Primary staging and follow-up in melanoma patients--monocenter evaluation of methods, costs and patient survival.

In a German cohort of 661 melanoma patients the performance, costs and survival benefits of staging methods (history and physical examination; chest X-ray; ultrasonography of the abdomen; high resolution sonography of the peripheral lymph nodes) were assessed at initial staging and during follow-up of stage I/II+III disease. At initial staging, 74% (23 out of 31) of synchronous metastases were first detected by physical examination followed by sonography of the lymph nodes revealing 16% (5 out of 31). Other imaging methods were less efficient (Chest X-ray: one out of 31; sonography of abdomen: two out of 31). Nearly 24% of all 127 first recurrences and 18% of 73 second recurrences developed in patients not participating in the follow-up programme. In follow-up patients detection of first or second recurrence were attributed to history and physical examination on a routine visit in 47 and 52% recurrences, respectively, and to routine imaging procedures in 21 and 17% of cases, respectively. Lymph node sonography was the most successful technical staging procedure indicating 13% of first relapses, but comprised 24% of total costs of follow-up in stage I/II. Routine imaging comprised nearly 50% of total costs for follow-up in stage I/II and in stage III. The mode of detecting a relapse ('patient vs. doctor-diagnosed' or 'symptomatic vs asymptomatic') did not significantly influence patients overall survival. Taken together, imaging procedures for routine follow-up in stage I/II and stage III melanoma patients were inefficient and not cost-efficient.