WHY ManSegTool?

UNDERSTANDING HOW THE BRAIN WORKS IS ARGUABLY ONE OF THE GREATEST CHALLENGES OF OUR TIME.  A FUNDAMENTAL LIMITATION FOR EXPLORING THE FUNCTION OF COMPLEX NEURAL CIRCUITS AND THEIR ALTERATIONS IN PATHOLOGICAL BRAIN PROCESSES IS OUR LACK OF KNOWLEDGE ON THE MICRO-ARCHITECTURE AND ORGANIZATION OF NEURONS WITHIN THE BRAIN.

WHILE BRAIN TISSUE IMAGING METHODS ARE CONSTANTLY IMPROVING THEIR PERFORMANCE, POWERFUL COMPUTATIONAL ALGORITHMS AND TOOLS TO SEGMENT NEURAL STRUCTURES ARE STILL LACKING, DESPITE THE ORGANIZATION OF COMPETITIONS IN THE FIELD (I.E. DIADEM CHALLENGE). TO DATE, MANUAL SEGMENTATION IS CONSIDERED NECESSARY AS A YARDSTICK AGAINST WHICH RECONSTRUCTION ALGORITHM CAN BE COMPARED.

MOREOVER, DURING THE DIADEM CHALLENGE A METRIC WAS ESTABLISHED TO EVALUATE THE GOODNESS OF SEGMENTATION, CAPTURING THE MOST CRITICAL ASPECTS IN NEURONAL SHAPE, BUT LEAVING OUT USEFUL INFORMATION (e.g. neurite thickness). HOWEVER, THE NEW ALGORITHMS AND TOOL CONTINUE TO BE VALIDATED IN DIFFERENT WAYS IN DIFFERENT METHODS.

TO OVERCOME THESE LIMITS, WE IMPLEMENTED ManSegTool TO FACILITATE THE MANUAL SEGMENTATION OF NEURONS FROM IMAGE STACKS.