The 2016 revision of the WHO classification of lymphoid neoplasms: The clinician's view

A Transfer Learning Radiomics Nomogram for Preoperative Prediction of Borrmann Type IV Gastric Cancer From Primary Gastric Lymphoma

Objective

This study aims to differentiate preoperative Borrmann type IV gastric cancer (GC) from primary gastric lymphoma (PGL) by transfer learning radiomics nomogram (TLRN) with whole slide images of GC as source domain data.

Materials and Methods

This study retrospectively enrolled 438 patients with histopathologic diagnoses of Borrmann type IV GC and PGL. They received CT examinations from three hospitals. Quantitative transfer learning features were extracted by the proposed transfer learning radiopathomic network and used to construct transfer learning radiomics signatures (TLRS). A TLRN, which integrates TLRS, clinical factors, and CT subjective findings, was developed by multivariate logistic regression. The diagnostic TLRN performance was assessed by clinical usefulness in the independent validation set.

Results

The TLRN was built by TLRS and a high enhanced serosa sign, which showed good agreement by the calibration curve. The TLRN performance was superior to the clinical model and TLRS. Its areas under the curve (AUC) were 0.958 (95% confidence interval [CI], 0.883–0.991), 0.867 (95% CI, 0.794–0.922), and 0.921 (95% CI, 0.860–0.960) in the internal and two external validation cohorts, respectively. Decision curve analysis (DCA) showed that the TLRN was better than any other model. TLRN has potential generalization ability, as shown in the stratification analysis.

Conclusions

The proposed TLRN based on gastric WSIs may help preoperatively differentiate PGL from Borrmann type IV GC.

Borrmann type IV gastric cancer, primary gastric lymphoma, transfer learning, whole slide image, deep learning.

Two-way detection of image features and immunolabeling of lymphoma cells with one-step microarray analysis

Detecting the number of pathological lymphoma cells and lymphocyte subtypes in blood is helpful for clinical diagnosis and typing of lymphoma. In the current study, cell type is identified by cell morphological features and immunolabeled lymphocyte subtypes. Red blood cells and leukocytes were separated using a microfluidic cell chip based on physical blood cell parameters, and leukocytes were identified using five characteristic parameters: energy variance, entropy variance, moment of inertia variance, color mean, and cell area individually. The number of red blood cells that could come into contact with the leukocyte membrane was ≤2 based on the microfluidic injection flow rate of microfluidic chips. Anti-CD3 and anti-CD19 antibodies were used for immunofluorescence staining of T-lymphocyte and B-lymphocyte surface antigens, respectively. The results suggested that the microfluidic assay could detect lymphocyte surface antigen markers and intact leukocytes. Therefore, we report a one-step microfluidic chip for classifying hematological lymphoma cells based on the physical parameters of cells, which can simultaneously measure the overall morphology of blood cells and immunolabeling of lymphocyte surface antigens in one step, solving the current problem of detecting subtypes of hematological lymphoma cells based on multiple methods and multi-step detection.