Deep learning-based classifier of diffuse large B-cell lymphoma cell-of-origin with clinical outcome
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Diffuse large B-cell lymphoma (DLBCL) is an aggressive form of non-Hodgkin lymphoma with poor response to R-CHOP therapy due to remarkable heterogeneity. Based on gene expression, DLBCL cases were divided into two subtypes, i.e. ABC and GCB, where ABC subtype is associated with poor outcomes. Due to its association with clinical outcome, this classification, also known as cell-of-origin (COO), is an efficient way to predict the response to R-CHOP therapy. Previous COO classification methods have some shortcomings, e.g. limited number of samples in the training dataset. These shortcomings challenge the robustness of methods and make it difficult to implicate these methods at clinical level. To overcome the shortcomings of previous methods, we developed a deep learning-based classifier model on a cohort of 381 DLBCL patients using expression data of 20 genes. We implemented multilayer perceptron (MLP) to train deep learning-based classifier, named MLP-COO. MLP-COO achieved accuracy of 99.70% and 94.70% on training and testing datasets, respectively, with 10-fold cross-validation. We also assessed its performance on an independent dataset of 294 DLBCL patients. On independent dataset, we achieved an accuracy of 95.90% with MCC of 0.917. To show its broader applicability, we used this classifier to predict the clinical outcome using survival data from two large cohorts of DLBCL patients. In survival analysis, MLP-COO recapitulates the survival probabilities of DLBCL patients based on their COO in both cohorts. We anticipate that MLP-COO model developed in this study will benefit in the accurate COO prediction of DLBCL patients and their clinical outcomes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it