Jie Tian

Development of a deep learning‐based nomogram for predicting lymph node metastasis in cervical cancer: A multicenter study

The pathological risk score: A new deep learning‐based signature for predicting survival in cervical cancer

AbstractPurposeTo develop and validate a deep learning‐based pathological risk score (RS) with an aim of predicting patients' prognosis to investigate the potential association between the information within the whole slide image (WSI) and cervical cancer prognosis.MethodsA total of 251 patients with the International Federation of Gynecology and Obstetrics (FIGO) Stage IA1–IIA2 cervical cancer who underwent surgery without any preoperative treatment were enrolled in this study. Both the clinical characteristics and WSI of each patient were collected. To construct a prognosis‐associate RS, high‐dimensional pathological features were extracted using a convolutional neural network with an autoencoder. With the score threshold selected by X‐tile, Kaplan–Meier survival analysis was applied to verify the prediction performance of RS in overall survival (OS) and disease‐free survival (DFS) in both the training and testing datasets, as well as different clinical subgroups.ResultsFor the OS and DFS prediction in the testing cohort, RS showed a Harrell's concordance index of higher than 0.700, while the areas under the curve (AUC) achieved up to 0.800 in the same cohort. Furthermore, Kaplan–Meier survival analysis demonstrated that RS was a potential prognostic factor, even in different datasets or subgroups. It could further distinguish the survival differences after clinicopathological risk stratification.ConclusionIn the present study, we developed an effective signature in cervical cancer for prognosis prediction and patients' stratification in OS and DFS.

Intraoperative near-infrared fluorescence imaging can identify pelvic nerves in patients with cervical cancer in real time during radical hysterectomy

Abstract Purpose Radical hysterectomy combined with pelvic lymphadenectomy is the standard treatment for early-stage cervical cancer, but unrecognized pelvic nerves are vulnerable to irreversible damage during surgery. This early clinical trial investigated the feasibility and safety of intraoperative near-infrared (NIR) fluorescence imaging (NIR-FI) with indocyanine green (ICG) for identifying pelvic nerves during radical hysterectomy for cervical cancer. Methods Sixty-six adults with cervical cancer were enrolled in this prospective, open-label, single-arm, single-center clinical trial. NIR-FI was performed in vivo to identify genitofemoral (GN), obturator (ON), and hypogastric (HN) nerves intraoperatively. The primary endpoint was the presence of fluorescence in pelvic nerves. Secondary endpoints were the ICG distribution in a nerve specimen and potential underlying causes of fluorescence emission in pelvic nerves. Results In total, 63 patients were analyzed. The ON was visualized bilaterally in 100% (63/63) of patients, with a mean fluorescence signal-to-background ratio (SBR) of 5.3±2.1. The GN was identified bilaterally in 93.7% (59/63) of patients and unilaterally in the remaining 4 patients, with a mean SBR of 4.1±1.9. The HN was identified bilaterally in 81.0% (51/63) of patients and unilaterally in 7.9% (5/63) of patients, with a mean SBR of 3.5±1.3. ICG fluorescence was detected in frozen sections of a nerve specimen, and was mainly distributed in axons. No ICG-related complications were observed. Conclusion This early clinical trial demonstrated the feasibility and safety of NIR-FI to visualize pelvic nerves intraoperatively. Thus, NIR-FI may help surgeons adjust surgical decision-making, avoid nerve damage, and improve surgical outcomes. Trial registration ClinicalTrials.gov NCT04224467

The Application of Real-Time Near-infrared Imaging in Gynecological Surgery

Removing in situ and metastasis lesions completely during gynecological surgery is central to reduce the recurrence and death, and the identification of lesions in traditional gynecological surgery often depends on the experience of surgeons. The identification of nerves is often needed in gynecological surgery, such as the obturator nerves in pelvic lymphadenectomy, and the pelvic autonomic nerves in nerve-sparing radical hysterectomy for cervical cancer. Nerve identification also relies heavily on the experience of surgeons. This project aims to realize the identification of lesions and nerves under the navigation of indocyanine green fluorescent imaging, and evaluate the accuracy of fluorescent imaging of lesions and the effectiveness of nerves identification by near-infrared imaging. This project may reduce the recurrence or death caused by residual lesions and postoperative dysfunction caused by nerves injury, thus, improve the survival rate and quality of life for patients with gynecological diseases.