Jun-Ho Choi, Dong-Wan Seo, Tae Jun Song, Do Hyun Park, Sang Soo Lee, Sung Koo Lee, and Myung-Hwan Kim. Gut and Liver 1999;0:. https://doi.org/10.5009/gnl19123

Mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN) account for a large portion of solitary pancreatic cystic neoplasms (PCN). In this study we implemented a convolutional neural network (CNN) model using ResNet50 to differentiate between MCN and SCN. The training data were collected retrospectively from 59 MCN and 49 SCN patients from two different hospitals. Data augmentation was used to enhance the size and quality of training datasets. Fine-tuning training approaches were utilized by adopting the pre-trained model from transfer learning while training selected layers. Testing of the network was conducted by varying the endoscopic ultrasonography (EUS) image sizes and positions to evaluate the network performance for differentiation. The proposed network model achieved up to 82.75% accuracy and a 0.88 (95% CI: 0.817–0.930) area under curve (AUC) score. The performance of the implemented deep learning networks in decision-making using only EUS images is comparable to that of traditional manual decision-making using EUS images along with supporting clinical information. Gradient-weighted class activation mapping (Grad-CAM) confirmed that the network model learned the features from the cyst region accurately. This study proves the feasibility of diagnosing MCN and SCN using a deep learning network model. Further improvement using more datasets is needed.

Min Keun Cho, Sung-Hoon Moon, Tae Jun Song, Raymond E. Kim, Dong Wook Oh, Do Hyun Park, Sang Soo Lee, Dong Wan Seo, Sung Koo Lee, and Myung-Hwan Kim. Gut and Liver 1999;0:. https://doi.org/10.5009/gnl17391

The clinical benefit and potential risks of conversion surgery after neoadjuvant chemotherapy (NACT) have not been fully investigated in patients with borderline resectable pancreatic cancer (BRPC) and locally advanced unresectable pancreatic cancer (LAPC). Therefore, this has been evaluated in a retrospective, prospective cohort-based analysis. Between October 2005 and April 2017, 135 patients (65 with BRPC and 70 with LAPC) received conversion surgery after NACT. Exploratory analysis to assess clinical outcomes in comparison with patients who underwent upfront surgery in the same time period (n = 359) was also conducted. NACT with gemcitabine-based regimens (including gemcitabine monotherapy, gemcitabine-capecitabine combination, and gemcitabine-erlotinib combination) was used in 69 patients (51%) and FOLFIRINOX in 66 patients (49%). The median overall survival (OS) and disease-free survival (DFS) from the time of surgery was 25.4 months (95% CI, 18.6–32.2 months) and 9.0 months (95% CI, 6.8–11.2 months), respectively. The median OS and progression-free survival from the initiation of NACT was 29.7 months (95% CI, 22.5–36.8 months) and 13.4 months (95% CI, 12.5–14.4 months), respectively. In the exploratory analysis, conversion surgery after NACT was associated with a better median OS and DFS than upfront surgery (vs. 17.1 months; 95% CI, 15.5–18.7 months; p = 0.001 and vs. 7.1 months; 95% CI, 6.4–7.8 months; p = 0.005, respectively). There was no difference in length of hospital stay between the two groups, and conversion surgery after NACT showed a significantly lower incidence of postoperative complications than upfront surgery (38% vs. 27%, p = 0.03). Conversion surgery after NACT is a feasible and effective therapeutic strategy for the treatment of patients with BRPC and LAPC. Further clinical trials investigating optimal therapeutic strategies for BRPC and LAPC are warranted.

Dongwook Oh, Tae Jun Song, Sung-Hoon Moon, Jin Hee Kim, Joo Nam Lee, Seung-Mo Hong, Joune Seup Lee, Seok Jung Jo, Dong Hui Cho, Do Hyun Park, Sang Soo Lee, Dong-Wan Seo, Sung Koo Lee, and Myung-Hwan Kim. Gut and Liver 1999;0:. https://doi.org/10.5009/gnl18429