Abstract-We present two real-time hidden Markov model-based systems for recognizing sentence-level continuous American Sign Language (ASL) using a single camera to track the user's unadorned hands. The first system observes the user from a desk mounted camera and achieves 92 percent word accuracy. The second system mounts the camera in a cap worn by the user and achieves 98 percent accuracy (97 percent with an unrestricted grammar). Both experiments use a 40-word lexicon.

Well-differentiated liposarcoma/atypical lipomatous tumor and dedifferentiated liposarcoma can be difficult to distinguish from benign lipomatous neoplasms and other high-grade sarcomas, respectively. Cytogenetics in these tumors has identified ring and giant chromosomes composed of 12q13-15 amplicons including the MDM2 gene. Identifying MDM2 amplification by fluorescence in situ hybridization may prove an adjunctive tool in the diagnosis of lipomatous neoplasms. Dual color fluorescence in situ hybridization employing a laboratorydeveloped BAC label probe cocktail specific for MDM2 (12q15) and a probe for the centromeric region of chromosome 12 (Abbott Molecular, DesPlaines, IL) was performed on formalin-fixed and paraffin-embedded tissue including whole sections from atypical lipomatous tumors (n ¼ 13), dedifferentiated liposarcomas (n ¼ 14), benign lipomatous tumors (n ¼ 30), and pleomorphic sarcoma, not otherwise specified (n ¼ 10), and a tissue microarray containing a variety of high-grade sarcomas (n ¼ 63). An MDM2/chromosome 12 ratio Z2.0 was considered amplified, o2.0 nonamplified, and cases displaying 42 signals of both probes and an MDM2 ratio o2.0 polysomic for chromosome 12. Of the well-differentiated and dedifferentiated liposarcomas, 100% showed amplification of MDM2. Chromosome 12 polysomy was noted in 89% of spindle cell/pleomorphic lipomas, while all angiolipomas and lipomas were nonamplified and eusomic. MDM2 amplification was observed in 40% of pleomorphic sarcomas and a small subset of high-grade sarcomas (3/63). MDM2/chromosome 12 fluorescence in situ hybridization is a sensitive and specific tool (both 100%) in evaluating low-grade lipomatous neoplasms. The specificity decreases in high-grade sarcomas, as MDM2 amplification was observed in a small portion of pleomorphic sarcomas and high-grade sarcomas other than dedifferentiated liposarcomas. Importantly, none of the benign lipomatous lesions were MDM2 amplified and even cells in areas of well-differentiated liposarcomas with minimal cytologic atypia were amplified, making the probe a valuable tool in the diagnosis of even limited biopsy samples of well-differentiated lipomatous neoplasms.

Well-differentiated liposarcoma/atypical lipomatous tumor can be difficult to differentiate from benign lipomatous tumors, especially on limited biopsy material. Adjunctive tests for MDM2 (murine double minute 2) have proven useful in whole-tissue sections; however, their utility has not been determined within the increasingly popular core needle biopsy. Herein, we compare the ability of MDM2 immunohistochemistry and MDM2 fluorescence in situ hybridization (FISH) to discriminate benign lipomatous tumors from welldifferentiated liposarcoma on core needle biopsies. Well-differentiated liposarcoma (n ¼ 17) and an assortment of benign lipomatous tumors (n ¼ 37), which had concurrent or previous core needle biopsies, and resection specimens were subjected to both MDM2 immunohistochemistry and MDM2 FISH on both whole-tissue sections and corresponding core needle biopsy sections. Percentage tumor cells positive for MDM2 by immunohistochemistry and an MDM2:CEP12 FISH ratio was calculated in each biopsy and resection specimen pair and the results were compared. MDM2 FISH had a higher sensitivity (100%) and specificity (100%) compared with MDM2 immunohistochemistry (65 and 89%) in core needle biopsies, respectively. In addition, MDM2 immunohistochemistry had a false-positive rate of 11%, compared to 0% with FISH. The average MDM2:CEP12 ratio was similar in the biopsy material compared with the whole-tissue sections in both welldifferentiated liposarcoma and the benign lipomatous tumor group of neoplasms. Detection of MDM2 amplification by FISH is a more sensitive and specific adjunctive test than MDM2 immunohistochemistry to differentiate well-differentiated liposarcoma from various benign lipomatous tumors, especially on limited tissue samples.

We propose a practical application of wearable computing and augmented reality which enhances the game of billiards. A vasaon algorathm as implemented whach operates in interactive-lime with the user to assisl planning and aiming. Probabilistic color models and symm.etry operations are used to localize th,c table, pockets and balls through a video camera near the user's e y e . Classification of the objects of interest is performed and each possible shot is ranked in order t o determine its relative usefulness. The system allows the user t o proceed through a regular pool gam,e wh,ile it automatically determines strategic shots. The resulting trajectories are rendered as graphical overlays on a head mounted live video display. The wearable video output and the computer vision system provide an integration of real and virtual environments which enhances the experieme of playing and learning the game of billiards without encumbering the player.