Search

Over the past decades, advances in surgical robotics have led to significant improvements in the care provided to women with gynecologic cancer. These advancements have occurred during a time when healthcare systems are changing their focus from a traditional fee for service model to value-based healthcare. The focus on value and risk sharing has led to institutions choosing quality outcomes over complications. This has been emphasized from the recent addition of the Institute for Healthcare Improvement modification of the Triple AIM to the Quadruple Aim: better care, better quality, better cost, and better experience. The added focus on better experience includes the patient and healthcare professional. This comes at an important time in medicine, where the implementation of electronic medical records (EMRs) such as Epic and Cerner has been met with a significant increase in physician burnout.

Cervical cancer is the second most common type of cancer for women. Existing screening programs for cervical cancer, such as Pap Smear, suffer from low sensitivity. Thus, many patients who are ill are not detected in the screening process. Using images of the cervix as an aid in cervical cancer screening has the potential to greatly improve sensitivity, and can be especially useful in resource-poor regions of the world. In this paper, we develop a data-driven computer algorithm for interpreting cervical images based on color and texture. We are able to obtain 74% sensitivity and 90% specificity when differentiating high-grade cervical lesions from low-grade lesions and normal tissue. On the same dataset, using Pap tests alone yields a sensitivity of 37% and specificity of 96%, and using HPV test alone gives a 57% sensitivity and 93% specificity. Furthermore, we develop a comprehensive algorithmic framework based on Multimodal Entity Coreference for combining various tests to perform disease classification and diagnosis. When integrating multiple tests, we adopt information gain and gradient-based approaches for learning the relative weights of different tests. In our evaluation, we present a novel algorithm that integrates cervical images, Pap, HPV, and patient age, which yields 83.21% sensitivity and 94.79% specificity, a statistically significant improvement over using any single source of information alone.

Hybrid simulators provide objective metrics for laparoscopic task performance. This study aimed to evaluate the correlation between hybrid simulator-generated metrics and content-valid outcome measures. Residents underwent training with a previously validated 5-task simulation model (5-TSM). The resident tasks included vessel clipping and dividing, lesion excision, loop appendectomy, mesh placement with tacks, and suture perforation with intracorporeal knot tying. After training, the residents were tested using the open module of a hybrid simulator (ProMIS) with previously validated passing scores. Content validity was defined as the extent to which outcome measures departed from clinical reality. Content-valid outcome measures (accuracy error, knot slippage, leak, operating time, tissue damage) were evaluated by two blinded raters. The hybrid simulator-generated metrics were path length and smoothness of movements. Values are given as means (standard deviation is not shown). Over 23 months, 20 residents underwent training with 5-TSM. Respectively, for tasks 1 to 5, the path length was 3,895, 3,472, 4,620, 2,408, and 9,089 mm; the smoothness (jerk) was 346, 455, 549, 264, and 910 cm/sec3; the accuracy error was 0.45, 2.20, 0.55, 0.87, and 0.20 mm; and the knot slippage was 5%. There were no leaks. The operating time, respectively, was 54, 61, 135, 43, and 130 s, and the tissue damage was 0, 0.28, 0, 0.8, and 0 mm. The interrater reliability was more than 0.80 for all the outcome measures except accuracy error (k = 0.52). There was correlation between path length and operating time (Spearman rho, 0.537–0.709; p

5582 Background: Sexually active adolescents have high rates of infection with human papilloma virus (HPV) and abnormal pap smears. They are considered a special population as they are likely to regress to normal cytology. The aim of our study was to identify factors associated with regression of cervical dysplasia in adolescents. Methods: We identified adolescent patients (aged 12–21 years) who had abnormal pap smears at the Center for Women's Medicine at Lehigh Valley Hospital in Allentown, PA, by CPT code from a database between Jan 2004 and Dec 2006. A chart review was performed to capture demographic data, cytology, smoking history, number of sexual partners, parity, race, contraceptive choice, use of barrier contraception. Chi-square analysis with logistic regression and multivariate analysis were used to identify factors associated with regression of cervical dysplasia. Results: Two-hundred two patients were identified. Mean age was 18.84 years (14–22 years). One hundred twenty-two (57.8%) were Hispanic, 71 (33.6%) Caucasian, and 16 (7.6%) Black. Fifty-two (24.6%) were pregnant at the time of diagnosis. Seventy-six (36%) were smokers. There were 125 (61.9%) cases of ASCUS, 33.7% (68 cases) LGSIL and 4.5% (9 cases) HGSIL on initial pap smear. One hundred eighteen (55.9%) patients had colposcopy, and of these, 32 (15.2%) had surgical intervention. Follow-up demonstrated that 72 (57.6%) patients had disease regression, 24 (19.2%) persistence and 29 (23.2%) progression. On multivariate analysis, patients who did not smoke were significantly more likely to show regression of cervical dysplasia on pap smear than women who smoked (OR 2.17, 95% CI 1.03–4.55, p = 0.039). Other factors were not statistically significant in predicting regression of cervical dysplasia. Conclusions: Adolescents who smoke were more likely to have persistent cervical dysplasia than non-smoking adolescents, putting smokers at an increased risk of advanced disease. We suggest that this subset have follow-up at shorter intervals and be enrolled in a smoking cessation program. No significant financial relationships to disclose.