Prediction of SARS-CoV-2-positivity from million-scale complete blood counts using machine learning

Unlocking the complete blood count as a risk stratification tool for breast cancer using machine learning: a large scale retrospective study

Optimizing early breast cancer (BC) detection requires effective risk assessment tools. This retrospective study from Brazil showcases the efficacy of machine learning in discerning complex patterns within routine blood tests, presenting a globally accessible and cost-effective approach for risk evaluation. We analyzed complete blood count (CBC) tests from 396,848 women aged 40-70, who underwent breast imaging or biopsies within six months after their CBC test. Of these, 2861 (0.72%) were identified as cases: 1882 with BC confirmed by anatomopathological tests, and 979 with highly suspicious imaging (BI-RADS 5). The remaining 393,987 participants (99.28%), with BI-RADS 1 or 2 results, were classified as controls. The database was divided into modeling (including training and validation) and testing sets based on diagnostic certainty. The testing set comprised cases confirmed by anatomopathology and controls cancer-free for 4.5-6.5 years post-CBC. Our ridge regression model, incorporating neutrophil-lymphocyte ratio, red blood cells, and age, achieved an AUC of 0.64 (95% CI 0.64-0.65). We also demonstrate that these results are slightly better than those from a boosting machine learning model, LightGBM, plus having the benefit of being fully interpretable. Using the probabilistic output from this model, we divided the study population into four risk groups: high, moderate, average, and low risk, which obtained relative ratios of BC of 1.99, 1.32, 1.02, and 0.42, respectively. The aim of this stratification was to streamline prioritization, potentially improving the early detection of breast cancer, particularly in resource-limited environments. As a risk stratification tool, this model offers the potential for personalized breast cancer screening by prioritizing women based on their individual risk, thereby indicating a shift from a broad population strategy.

Serum Thyrotropin and Triiodothyronine Levels in Levothyroxine-treated Patients

CONTEXT

Small adjustments in levothyroxine (LT4) dose do not appear to provide clinical benefit despite changes in thyrotropin (TSH) levels within the reference range. We hypothesize that the accompanying changes in serum total triiodothyronine (T3) levels do not reflect the magnitude of the changes in serum TSH.

OBJECTIVE

This work aims to characterize the relationships of serum free thyroxine (FT4) vs T3, FT4 vs TSH, and FT4 vs the T3/FT4 ratio.

METHODS

This cross-sectional, observational study comprised 9850 participants aged 18 years and older treated with LT4 from a large clinical database from January 1, 2009, to December 31, 2019. Patients had been treated with LT4, subdivided by serum FT4 level. Main outcome measures included model fitting of the relationships between serum FT4 vs TSH, FT4 vs T3, and FT4 vs T3/FT4. Mean and median values of TSH, T3, and T3/FT4 were calculated.

RESULTS

The relationships T3 vs FT4 and TSH vs FT4 were both complex and best represented by distinct, segmented regression models. Increasing FT4 levels were linearly associated with T3 levels until an inflection point at an FT4 level of 0.7 ng/dL, after which a flattening of the slope was observed following a convex quadratic curve. In contrast, increasing FT4 levels were associated with steep declines in TSH following 2 negative sigmoid curves. The FT4 vs T3/FT4 relationship was fit to an asymptotic regression curve supporting less T4 to T3 activation at higher FT4 levels.

CONCLUSION

In LT4-treated patients, the relationships between serum FT4 vs TSH and FT4 vs T3 across a range of FT4 levels are disproportionate. As a result, dose changes in LT4 that robustly modify serum FT4 and TSH values may only minimally affect serum T3 levels and result in no significant clinical benefit.

One-stage and lightweight CNN detection approach with attention: Application to WBC detection of microscopic images

White blood cell (WBC) detection in microscopic images is indispensable in medical diagnostics; however, this work, based on manual checking, is time-consuming, labor-intensive, and easily results in errors. Using object detectors for WBCs with deep convolutional neural networks can be regarded as a feasible solution. In this paper, to improve the examination precision and efficiency, a one-stage and lightweight CNN detector with an attention mechanism for detecting microscopic WBC images, and a white blood cell detection vision system are proposed. The method integrates different optimizing strategies to strengthen the feature extraction capability through the combination of an improved residual convolution module, hybrid spatial pyramid pooling module, improved coordinate attention mechanism, efficient intersection over union (EIOU) loss and Mish activation function. Extensive ablation and contrast experiments on the latest public Raabin-WBC dataset verify the effectiveness and robustness of the proposed detector for achieving a better overall detection performance. It is also more efficient than other existing studies for blood cell detection on two additional classic public BCCD and LISC datasets. The novel detection approach is significant and flexible for medical technicians to use for blood cell microscopic examination in clinical practice.

Cardiovascular and Renal Comorbidities Included into Neural Networks Predict the Outcome in COVID-19 Patients Admitted to an Intensive Care Unit: Three-Center, Cross-Validation, Age- and Sex-Matched Study

Here, we performed a multicenter, age- and sex-matched study to compare the efficiency of various machine learning algorithms in the prediction of COVID-19 fatal outcomes and to develop sensitive, specific, and robust artificial intelligence tools for the prompt triage of patients with severe COVID-19 in the intensive care unit setting. In a challenge against other established machine learning algorithms (decision trees, random forests, extra trees, neural networks, k-nearest neighbors, and gradient boosting: XGBoost, LightGBM, and CatBoost) and multivariate logistic regression as a reference, neural networks demonstrated the highest sensitivity, sufficient specificity, and excellent robustness. Further, neural networks based on coronary artery disease/chronic heart failure, stage 3-5 chronic kidney disease, blood urea nitrogen, and C-reactive protein as the predictors exceeded 90% sensitivity and 80% specificity, reaching AUROC of 0.866 at primary cross-validation and 0.849 at secondary cross-validation on virtual samples generated by the bootstrapping procedure. These results underscore the impact of cardiovascular and renal comorbidities in the context of thrombotic complications characteristic of severe COVID-19. As aforementioned predictors can be obtained from the case histories or are inexpensive to be measured at admission to the intensive care unit, we suggest this predictor composition is useful for the triage of critically ill COVID-19 patients.

Prognosing the risk of COVID-19 death through a machine learning-based routine blood panel: A retrospective study in Brazil

Background:

Despite an extensive network of primary care availability, Brazil has suffered profoundly during the COVID-19 pandemic, experiencing the greatest sanitary collapse in its history. Thus, it is important to understand phenotype risk factors for SARS-CoV-2 infection severity in the Brazilian population in order to provide novel insights into the pathogenesis of the disease.

Objective:

This study proposes to predict the risk of COVID-19 death through machine learning, using blood biomarkers data from patients admitted to two large hospitals in Brazil.

Methods:

We retrospectively collected blood biomarkers data in a 24-h time window from 6,979 patients with COVID-19 confirmed by positive RT-PCR admitted to two large hospitals in Brazil, of whom 291 (4.2%) died and 6,688 (95.8%) were discharged. We then developed a large-scale exploration of risk models to predict the probability of COVID-19 severity, finally choosing the best performing model regarding the average AUROC. To improve generalizability, for each model five different testing scenarios were conducted, including two external validations.

Results:

We developed a machine learning-based panel composed of parameters extracted from the complete blood count (lymphocytes, MCV, platelets and RDW), in addition to C-Reactive Protein, which yielded an average AUROC of 0.91 ± 0.01 to predict death by COVID-19 confirmed by positive RT-PCR within a 24-h window.

Conclusion:

Our study suggests that routine laboratory variables could be useful to identify COVID-19 patients under higher risk of death using machine learning. Further studies are needed for validating the model in other populations and contexts, since the natural history of SARS-CoV-2 infection and its consequences on the hematopoietic system and other organs is still quite recent.