Sreenath Madathil

Trajectories of body fatness in adulthood and the risk of ovarian cancer.

While excess body fatness in older adulthood has been linked to ovarian cancer, the influence of changes in body fatness over time is unclear. This study examined the association between adulthood trajectories of body mass index (BMI), a proxy for body fatness, and ovarian cancer. In a population-based case-control study (440 cases, 820 controls), we used a group-based trajectory approach to identify BMI trajectories from age 20-70. Using unconditional logistic regression, we estimated adjusted odds ratios (aOR) and 95 % confidence intervals (95 % CI) for the associations between the estimated trajectories and ovarian cancer. We identified three distinct BMI trajectories: a normal-stable trajectory, a normal-to-overweight trajectory and an overweight-to-obese trajectory, which included 63.2 %, 31.0 % and 6.8 % of the population, respectively. Multivariable aORs suggested that participants with normal weight at the onset of adulthood who became overweight over their adulthood time did not differ in their risk of ovarian cancer compared to those who maintained a normal weight throughout adulthood (aOR (95 %CI): 0.89 (0.69-1.16)). Among those in the overweight-to-obese trajectory, the aOR (95 %CI) was 1.45 (0.87-2.43), and thus in the direction of an increased ovarian cancer risk compared to those who maintained a normal weight. Our findings underscore the need for further research to clarify the role of body fatness across the lifetime in the etiology of ovarian cancer.

A multimodal deep learning model for cervical pre-cancers and cancers prediction: Development and internal validation study

The current cervical cancer screening and diagnosis have limitations due to their subjectivity and lack of reproducibility. We describe the development of a deep learning (DL)-based diagnostic risk prediction model and evaluate its potential for clinical impact. We developed and internally validated a DL model which accommodates both clinical data and colposcopy images in predicting the patients CIN2+ status using a retrospective cohort of 6356 cases of LEEP-conization/cone-biopsy (gold-standard diagnosis) following an abnormal screening result. The overall performance, discrimination, and calibration of the model were compared to expert clinician's colposcopic impression. The potential for clinical impact was assessed with rate of unnecessary conizations that could be avoided by using our model. The model combining clinical history and colposcopy images demonstrated superior performance prediction of CIN2+(AUC-ROC = 95.3 %, accuracy = 90.8 %, PPV = 94.1 %, NPV = 87.9 %) and better calibration compared to models that used image or clinical history data alone and outperformed clinician's colposcopic impressions. Moreover, if a decision threshold of 10 % is applied to the predicted probability from this model to recommend conization, up to 35 % of conizations could be avoided without missing any true CIN2+ cases. We present a novel DL model to predict cervical neoplasia with potential for reducing unnecessary conization. External validation studies are warranted for assessing generalizability.