Deep learning-based prediction of post-pancreaticoduodenectomy pancreatic fistula.

Postoperative pancreatic fistula is a life-threatening complication with an unmet need for accurate prediction. This study was aimed to develop preoperative artificial intelligence-based prediction models. Patients who underwent pancreaticoduodenectomy were enrolled and stratified into model development and validation sets by surgery between 2016 and 2017 or in 2018, respectively. Machine learning models based on clinical and body composition data, and deep learning models based on computed tomographic data, were developed, combined by ensemble voting, and final models were selected comparison with earlier model. Among the 1333 participants (training, n = 881; test, n = 452), postoperative pancreatic fistula occurred in 421 (47.8%) and 134 (31.8%) and clinically relevant postoperative pancreatic fistula occurred in 59 (6.7%) and 27 (6.0%) participants in the training and test datasets, respectively. In the test dataset, the area under the receiver operating curve [AUC (95% confidence interval)] of the selected preoperative model for predicting all and clinically relevant postoperative pancreatic fistula was 0.75 (0.71-0.80) and 0.68 (0.58-0.78). The ensemble model showed better predictive performance than the individual ML and DL models.

Preoperative data-based deep learning model for predicting postoperative survival in pancreatic cancer patients.

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis even after curative resection. A deep learning-based stratification of postoperative survival in the preoperative setting may aid the treatment decisions for improving prognosis. This study was aimed to develop a deep learning model based on preoperative data for predicting postoperative survival. METHODS: The patients who underwent surgery for PDAC between January 2014 and May 2015. Clinical data-based machine learning models and computed tomography (CT) data-based deep learning models were developed separately, and ensemble learning was utilized to combine two models. The primary outcomes were the prediction of 2-year overall survival (OS) and 1-year recurrence-free survival (RFS). The model's performance was measured by area under the receiver operating curve (AUC) and was compared with that of American Joint Committee on Cancer (AJCC) 8th stage. RESULTS: The median OS and RFS were 23 and 10 months in training dataset (n = 229), and 22 and 11 months in test dataset (n = 53), respectively. The AUC of the ensemble model for predicting 2-year OS and 1-year RFS in the test dataset was 0.76 and 0.74, respectively. The performance of the ensemble model was comparable to that of the AJCC in predicting 2-year OS (AUC, 0.67; P = 0.35) and superior to the AJCC in predicting 1-year RFS (AUC, 0.54; P = 0.049). CONCLUSION: Our ensemble model based on routine preoperative variables showed good performance for predicting prognosis for PDAC patients after surgery.

Label-Free SERS Analysis of Urine Using a 3D-Stacked AgNW-Glass Fiber Filter Sensor for the Diagnosis of Pancreatic Cancer and Prostate Cancer.

Metabolomics shows tremendous potential for the early diagnosis and screening of cancer. For clinical application as an effective diagnostic tool, however, improved analytical methods for complex biological fluids are required. Here, we developed a reliable rapid urine analysis system based on surface-enhanced Raman spectroscopy (SERS) using 3D-stacked silver nanowires (AgNWs) on a glass fiber filter (GFF) sensor and applied it to the diagnosis of pancreatic cancer and prostate cancer. Urine samples were pretreated with centrifugation to remove large debris and with calcium ion addition to improve the binding of metabolites to AgNWs. The label-free urine-SERS detection using the AgNW-GFF SERS sensor showed different spectral patterns and distinguishable specific peaks in three groups: normal control (n = 30), pancreatic cancer (n = 22), and prostate cancer (n = 22). Multivariate analyses of SERS spectra using unsupervised principal component analysis and supervised orthogonal partial least-squares discriminant analysis showed excellent discrimination between the pancreatic cancer group and the prostate cancer group as well as between the normal control group and the combined cancer groups. The results demonstrate the great potential of the urine-SERS analysis system using the AgNW-GFF SERS sensor for the noninvasive diagnosis and screening of cancers.

Highly sensitive and repeatable DNA-SERS detection system using silver nanowires-glass fiber filter substrate.

This paper describes a new simple DNA detection method based on surface-enhanced Raman scattering (SERS) technology using a silver nanowire stacked-glass fiber filter substrate. In this system, DNA-intercalating dye (EVAGreen) and reference dye (ROX) are used together to improve the repeatability and reliability of the SERS signals. We found that the SERS signal of EVAGreen was reduced by intercalation into DNA amplicons of a polymerase chain reaction on the silver nanowire stacked-glass fiber filter substrate, whereas that of ROX stayed constant. The DNA amplicons could be quantified by correcting the EVAGreen-specific SERS signal intensity with the ROX-specific SERS signal intensity. Multivariate analysis by partial least square methods was also successfully performed. And we further applied it to loop-mediated isothermal amplification with potential use for on-site diagnostics. The sensitivities of the DNA-SERS detection showed about 100 times higher than those of conventional fluorescence-based detection methods. The DNA-SERS detection method can be applied to various isothermal amplification methods, which is expected to realize on-site molecular diagnostics with high sensitivity, repeatability, simplicity, affordability, and convenience.