Spectroscopic profiling-based geographic herb identification by neural network with random weights.

Daodi medicinal material plays an important role in traditional Chinese medicine (TCM). This study researches and validates the NNRW (neural network with random weights) model on spectroscopic profiling data for geographical origin identification. NNRW is a special neural network model that does not require an iterative training process. It has been proved effective in various resource-limited data-driven applications. However, whether NNRW works for spectroscopic profiling data remains to be explored. In this study, the Raman and UV (ultraviolet) profiling data of 160 radix astragali samples from four geographic regions are trained and evaluated by four classification models, i.e., NNRW, MLP (multi-layer perceptron), SVM (support vector machine), and DTC (decision tree classifier). Their validation accuracies are 96.3%, 98.0%, 98.4%, and 92.8% respectively. The training/fitting times are 0.372 ms (milli-seconds), 57.9 ms, 2.033 ms, and 3.351 ms, respectively. This study shows that NNRW has a significant training time cut while keeping a high prediction accuracy, and it is a promising solution to resource-limited edge computing applications.

Surgical strategies for treatment of T1b gallbladder cancers diagnosed intraoperatively or postoperatively / 北京大学学报(医学版)

Objective:To explore the surgical strategies for the treatment of T1 b gallbladder cancer patients diagnosed intraoperatively or postoperatively.Methods:A retrospective analysis of 42 patients with T1 b gallbladder cancers was performed.There were 14 patients diagnosed intraoperatively and 28 patients diagnosed postoperatively.The reevaluations of T stages were conducted in the 28 T1 b gallbladder cancer patients diagnosed postoperatively by the professional pathologist.After T stage reevaluation,25 confirmed T1 b patients with complete follow-up data were divided into simple cholecystectomy group and radical resection group,and the clinicopathologic characteristics between the two groups were analyzed.Results:Only 2 of the 14 T1 b gallbladder cancer patients diagnosed by the intraoperative frozen specimen proved to be T1b on postoperative paraffin pathology,and for the remaining 13 patients,T2 was in 11 patients,and T3 in one patient.The rate of misdiagnosis was 85.7％ by the intraoperative frozen specimens,postoperative T stages were equal or higher than intraoperative T stages.Two of the 28 postoperatively diagnosed T1b patients were proved to be T2 after reevaluation,the rate of misdiagnosis was 7.1％,the reevaluated T stages were equal to or higher than the previous stages.Twenty-five confirmed T1 b gallbladder cancer patients had complete follow-up data,11 of whom underwent simple cholecystectomy and the remaining 14 radical resections.No patient had vessel or perineural invasion on pathology in the 25 confirmed T1b patients.Metastasis was absent in all the 30 lymph nodes examined,which achieved from 14 patients with radical resection.The survival rate after simple cholecystectomy was comparable to that after radical resection (P =0.361).Only one patient with radical resection had abdominal cavity implantation relapse,who received gallbladder compression during operation and 2 years later died from metastasis.Conclusion:Intraoperatively diagnosed T1 b gallbladder cancer should receive radical resection.Reevaluation of the T stage is necessary and the initial step for postoperative diagnosed T1b gallbladder cancer patients.The pros and cons of radical surgery for definitive T1b patients should be carefully evaluated,and systemic chemotherapy is recommended for those with bile spillover.