Automated polar ice thickness estimation from radar imagery.

This paper focuses on automating the task of estimating Polar ice thickness from airborne radar data acquired over Greenland and Antarctica. This process involves the identification and accurate selection of the ice sheet's surface location and interface between the ice sheet and the underlying bedrock for each measurement. Identifying the surface and bedrock locations in the radar imagery enables the computation of ice sheet thickness, which is important for the study of ice sheets, their volume, and how they may contribute to global climate change. The time-consuming manual approach requires sparse hand-selection of surface and bedrock interfaces by several human experts, and interpolating between the selections to save time. Two primary methods have been studied in this paper, namely, edge-based and active contour. Results are evaluated and presented using the metrics of time requirements and accuracy. Automated ice thickness estimation results from 2006 and 2007 Greenland field campaigns illustrate that the edge-based approach offers faster processing (seconds compared to minutes), but suffers from a lack of continuity and smoothness aspects that active contours provide. The active contour approach is more accurate when compared to ground truth selections provided by human experts, and has proven to be more robust to image artifacts. It is shown that both techniques offer advantages which could be integrated to yield a more effective system.

Current clinical practices in stroke rehabilitation: regional pilot survey.

This study was aimed at understanding the current physical and occupational therapy practices in stroke rehabilitation in the Midwest. The insights gained from this pilot study will be used in a future study aimed at understanding stroke rehabilitation practices across the nation. Researchers and clinicians in the field of stroke rehabilitation were interviewed, and past studies in the literature were analyzed. Through these activities, we developed a 37-item questionnaire that was sent to occupational and physical therapists practicing in Kansas and Missouri who focus on the care of people who have had a stroke (n = 320). A total of 107 respondents returned a com pleted questionnaire, which gives a response rate of about 36%. The majority of respondents had more than 12 years of experience treating patients with stroke. Consensus of 70% or more was found for 80% of the items. The preferred approaches for the rehabilitation of people who have had a stroke are the Bobath and Brunnstrom methods, which are being used by 93% and 85% of the physical and occupational therapists, respectively. Even though some variability existed in certain parts of the survey, in general clinicians agreed on different treatment approaches in issues dealing with muscle tone, weakness, and limited range of motion in stroke rehabilitation. Some newer treatment approaches that have been proven to be effective are practiced only by a minority of clinicians. The uncertainty among clinicians in some sections of the survey reveals that more evidence on clinical approaches is needed to ensure efficacious treatments.

Artificial intelligence techniques for automatic screening of amblyogenic factors.

PURPOSE: To develop a low-cost automated video system to effectively screen children aged 6 months to 6 years for amblyogenic factors. METHODS: In 1994 one of the authors (G.C.) described video vision development assessment, a digitizable analog video-based system combining Brückner pupil red reflex imaging and eccentric photorefraction to screen young children for amblyogenic factors. The images were analyzed manually with this system. We automated the capture of digital video frames and pupil images and applied computer vision and artificial intelligence to analyze and interpret results. The artificial intelligence systems were evaluated by a tenfold testing method. RESULTS: The best system was the decision tree learning approach, which had an accuracy of 77%, compared to the "gold standard" specialist examination with a "refer/do not refer" decision. Criteria for referral were strabismus, including microtropia, and refractive errors and anisometropia considered to be amblyogenic. Eighty-two percent of strabismic individuals were correctly identified. High refractive errors were also correctly identified and referred 90% of the time, as well as significant anisometropia. The program was less correct in identifying more moderate refractive errors, below +5 and less than -7. CONCLUSIONS: Although we are pursuing a variety of avenues to improve the accuracy of the automated analysis, the program in its present form provides acceptable cost benefits for detecting ambylogenic factors in children aged 6 months to 6 years.