Probability calibration-based prediction of recurrence rate in patients with diffuse large B-cell lymphoma

Application of machine learning in the management of lymphoma: Current practice and future prospects

In the past decade, digitization of medical records and multiomics data analysis in lymphoma has led to the accessibility of high-dimensional records. The digitization of medical records, the visualization of extensive volume data extracted from medical images, and the integration of multiomics methods into clinical decision-making have produced many datasets. As a promising auxiliary tool, machine learning (ML) intends to extract homologous features in large-scale data sets and encode them into various patterns to complete complicated tasks. At present, artificial intelligence and digital mining have shown promising prospects in the field of lymphoma pathological image analysis. The paradigm shift from qualitative analysis to quantitative analysis makes the pathological diagnosis more intelligent and the results more accurate and objective. ML can promote accurate lymphoma diagnosis and provide patients with prognostic information and more individualized treatment options. Based on the above, this comprehensive review of the general workflow of ML highlights recent advances in ML techniques in the diagnosis, treatment, and prognosis of lymphoma, and clarifies the boundedness and future orientation of the ML technique in the clinical practice of lymphoma.

Construction and Validation of a Novel Nomogram for Predicting the Recurrence of Diffuse Large B Cell Lymphoma Treated with R-CHOP

Purpose

To explore recurrence-risk factors of diffuse large B cell lymphoma (DLBCL) and construct a risk nomogram for predicting recurrence.

Patients and Methods

A retrospective analysis was performed on 228 DLBCL patients who achieved complete remission after R-CHOP treatment between January 2015 and December 2019. Univariate and multivariate analyses were applied to identify recurrence-related risk factors from the pretreatment evaluation factors covering patients' demographic characteristics, clinical manifestations, serological indicators, pathological and immunohistochemical results. A nomogram was developed based on the above results and validated by the concordance index (C-index), the receiver operating characteristic (ROC) curve, and the calibration curve.

Results

The training and validation cohorts consisted of 160 and 68 patients (randomized by 7:3). Of the whole cohort, 50 of 228 (21.9%) cases recurred during follow-up. Three recurrence-risk factors including BCL2 expression (P = 0.027), CD10 expression (P = 0.021), LDH level (P = 0.004) were identified from multivariate analysis and entered the final nomogram. The C-index of the nomogram was 0.815 in training cohort and 0.797 in the validation cohort, higher than that of IPI system (0.699) and NCCN-IPI system (0.709). And the 1-year, 2-year, 3-year, and 4-year areas under ROC (AUC) were 0.812, 0.850, 0.837, and 0.801, respectively. The calibration curves also showed a good discrimination capability and accuracy.

Conclusion

The novel nomogram incorporating the three independent risk factors (BCL2 expression, CD10 expression and LDH level) provided a valuable tool for predicting DLBCL recurrence.

Deep learning methods may not outperform other machine learning methods on analyzing genomic studies

Deep Learning (DL) has been broadly applied to solve big data problems in biomedical fields, which is most successful in image processing. Recently, many DL methods have been applied to analyze genomic studies. However, genomic data usually has too small a sample size to fit a complex network. They do not have common structural patterns like images to utilize pre-trained networks or take advantage of convolution layers. The concern of overusing DL methods motivates us to evaluate DL methods' performance versus popular non-deep Machine Learning (ML) methods for analyzing genomic data with a wide range of sample sizes. In this paper, we conduct a benchmark study using the UK Biobank data and its many random subsets with different sample sizes. The original UK Biobank data has about 500k participants. Each patient has comprehensive patient characteristics, disease histories, and genomic information, i.e., the genotypes of millions of Single-Nucleotide Polymorphism (SNPs). We are interested in predicting the risk of three lung diseases: asthma, COPD, and lung cancer. There are 205,238 participants have recorded disease outcomes for these three diseases. Five prediction models are investigated in this benchmark study, including three non-deep machine learning methods (Elastic Net, XGBoost, and SVM) and two deep learning methods (DNN and LSTM). Besides the most popular performance metrics, such as the F1-score, we promote the hit curve, a visual tool to describe the performance of predicting rare events. We discovered that DL methods frequently fail to outperform non-deep ML in analyzing genomic data, even in large datasets with over 200k samples. The experiment results suggest not overusing DL methods in genomic studies, even with biobank-level sample sizes. The performance differences between DL and non-deep ML decrease as the sample size of data increases. This suggests when the sample size of data is significant, further increasing sample sizes leads to more performance gain in DL methods. Hence, DL methods could be better if we analyze genomic data bigger than this study.