No abstract

No abstract

No abstract

BackgroundComprehensive capture of Adverse Events (AEs) is crucial for monitoring for side effects of a therapy while assessing efficacy. For cancer studies, the National Cancer Institute has developed the Common Terminology Criteria for Adverse Events (CTCAE) as a required standard for recording attributes and grading AEs. The AE assessments should be part of the Electronic Health Record (EHR) system; yet, due to patient-centric EHR design and implementation, many EHR's don't provide straightforward functions to assess ongoing AEs to indicate a resolution or a grade change for clinical trials.MethodsAt UAMS, we have implemented a standards-based Adverse Event Reporting System (AERS) that is integrated with the Epic EHR and other research systems to track new and existing AEs, including automated lab result grading in a regulatory compliant manner. Within a patient's chart, providers can launch AERS, which opens the patient's ongoing AEs as default and allows providers to assess (resolution/ongoing) existing AEs. In another tab, it allows providers to create a new AE. Also, we have separated symptoms from diagnoses in the CTCAE to minimize inaccurate designation of the clinical observations. Upon completion of assessments, a physician would submit the AEs to the EHR via a Health Level 7 (HL7) message and then to other systems utilizing a Representational State Transfer Web Service.ConclusionsAERS currently supports CTCAE version 3 and 4 with more than 65 cancer studies and 350 patients on those studies. This type of standard integrated into the EHR aids in research and data sharing in a compliant, efficient, and safe manner.

We've been developing a Chinese OCR engine for handwritten Chinese scripts. Curre ntly, our OCR engine supports a vocab ulary of 4616 characters which include 4516 simplifi ed Chinese characters in GB23 12-80, 62 alphanumeric characters, 38 punctu ation marks and symbols. By using 1,384,800 character samples to train our recognizer, an averaged character recognition accuracy of 96.34% is achieved on a testing set of 1,025,535 character samples.An arguably best Chinese OCR product on the market achieves an accuracy of 94.07% for the recognizable Chinese characters in the above testing set. In this paper, we describe key techniques used in our recognizer that contnb ute to the high recognition accuracy, namely the use of Gabor features and their spatial derivatives as raw features. the use ofLDA for feature extraction and dimension reduction, the use of CDHMMs for modeling Chinese characters along both horizontal and vertical directions, and the use of mini mum classification error as a criterion for model training.