MétaCan
Menu
← Back to cohort
Record W4376224490 · doi:10.2196/45455

Epidemiology and a Predictive Model of Prognosis Index Based on Machine Learning in Primary Breast Lymphoma: Population-Based Study

2023· article· en· W4376224490 on OpenAlexvenueno aff
Yushuai Yu, Zelin Xu, Tinglei Shao, Kaiyan Huang, Ruiliang Chen, Xiaoqin Yu, Jie Zhang, Hui Han, Chuangui Song

Bibliographic record

VenueJMIR Public Health and Surveillance · 2023
Typearticle
Languageen
FieldMedicine
TopicLymphoma Diagnosis and Treatment
Canadian institutionsnot available
Fundersnot available
KeywordsEpidemiologyMedicineLymphomaPopulationIndex (typography)OncologyMedical physicsComputer scienceInternal medicineEnvironmental healthWorld Wide Web

Abstract

fetched live from OpenAlex

BACKGROUND: Primary breast lymphoma (PBL) is a rare disease whose epidemiological features, treatment principles, and factors used for the patients' prognosis remain controversial. OBJECTIVE: The aim of this study was to explore the epidemiology of PBL and to develop a better model based on machine learning to predict the prognosis for patients with primary breast lymphoma. METHODS: The annual incidence of PBL was extracted from the surveillance, epidemiology, and end results database between 1975 and 2019 to examine disease occurrence trends using Joinpoint software (version 4.9; National Cancer Institute). We enrolled data from 1251 female patients with primary breast lymphoma from the surveillance, epidemiology, and end results database for survival analysis. Univariable and multivariable analyses were performed to explore independent prognostic factors for overall survival and disease-specific survival of patients with primary breast lymphoma. Eight machine learning algorithms were developed to predict the 5-year survival of patients with primary breast lymphoma. RESULTS: The overall incidence of PBL increased drastically between 1975 and 2004, followed by a significant downward trend in incidence around 2004, with an average annual percent change (AAPC) of -0.8 (95% CI -1.1 to -0.6). Disparities in trends of PBL exist by age and race. The AAPC of the 65 years or older cohort was about 1.2 higher than that for the younger than 65 years cohort. The AAPC of White patients is 0.9 (95% CI 0.0-1.8), while that of Black patients was significantly higher at 2.1 (95% CI -2.5 to 6.9). We also identified that the risk of death from PBL is multifactorial and includes patient factors and treatment factors. Survival analysis revealed that the patients diagnosed between 2007 and 2015 had a significant risk reduction of mortality compared to those diagnosed between 1983 and 1990. The gradient booster model outperforms other models, with 0.752 for sensitivity and 0.817 for area under the curve. The important features established with the gradient booster model were the year of diagnosis, age, histologic type, and primary site, which were the 4 most relevant variables to explain 5-year survival status. CONCLUSIONS: The incidence of PBL started demonstrating a tendency to decrease after 2004, which varied by age and race. In recent years, the prognosis of patients with primary breast lymphoma has been remarkably improved. The gradient booster model had a promising performance. This model can help clinicians identify the early prognosis of patients with primary breast lymphoma and therefore improve the clinical outcome by changing management strategies and patient health care.

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.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.007
metaresearch head score (Gemma)0.020
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.007
Threshold uncertainty score0.035

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0070.020
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0020.001
Science and technology studies0.0000.000
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0020.000

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.

Opus teacher head0.040
GPT teacher head0.318
Teacher spread0.279 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations5
Published2023
Admission routes1
Has abstractyes

Explore more

Same venueJMIR Public Health and Surveillance→Same topicLymphoma Diagnosis and Treatment→French-language works237,207→