An automated COVID-19 triage pipeline using artificial intelligence based on chest radiographs and clinical data

While COVID-19 diagnosis and prognosis artificial intelligence models exist, very few can be implemented for practical use given their high risk of bias. We aimed to develop a diagnosis model that addresses notable shortcomings of prior studies, integrating it into a fully automated triage pipeline that examines chest radiographs for the presence, severity, and progression of COVID-19 pneumonia. Scans were collected using the DICOM Image Analysis and Archive, a system that communicates with a hospital’s image repository. The authors collected over 6,500 non-public chest X-rays comprising diverse COVID-19 severities, along with radiology reports and RT-PCR data. The authors provisioned one internally held-out and two external test sets to assess model generalizability and compare performance to traditional radiologist interpretation. The pipeline was evaluated on a prospective cohort of 80 radiographs, reporting a 95% diagnostic accuracy. The study mitigates bias in AI model development and demonstrates the value of an end-to-end COVID-19 triage platform.

Algorithmic Fairness and Bias Mitigation for Clinical Machine Learning: Insights from Rapid COVID-19 Diagnosis by Adversarial Learning

Machine learning is becoming increasingly promi- nent in healthcare. Although its benefits are clear, growing attention is being given to how machine learning may exacerbate existing biases and disparities. In this study, we introduce an adversarial training framework that is capable of mitigating biases that may have been acquired through data collection or magnified during model development. For example, if one class is over-presented or errors/inconsistencies in practice are reflected in the training data, then a model can be biased by these. To evaluate our adversarial training framework, we used the statistical definition of equalized odds. We evaluated our model for the task of rapidly predicting COVID-19 for patients presenting to hospital emergency departments, and aimed to mitigate regional (hospital) and ethnic biases present. We trained our framework on a large, real-world COVID-19 dataset and demonstrated that adversarial training demonstrably improves outcome fairness (with respect to equalized odds), while still achieving clinically-effective screening performances (NPV>0.98). We compared our method to the benchmark set by related previous work, and performed prospective and external validation on four independent hospital cohorts. Our method can be generalized to any outcomes, models, and definitions of fairness.

Effects of buprenorphine on model development in an adjuvant-induced monoarthritis rat model

Complete Freund’s adjuvant (CFA)-induced arthritis in rats is a common animal model for studying chronic inflammatory pain. However, modelling of the disease is associated with unnecessary pain and impaired animal wellbeing, particularly in the immediate post-induction phase. Few attempts have been made to counteract these adverse effects with analgesics. The present study investigated the effect of buprenorphine on animal welfare, pain-related behaviour and model-specific parameters during the disease progression in a rat model of CFA-induced monoarthritis. The aim was to reduce or eliminate unnecessary pain in this model, in order to improve animal welfare and to avoid suffering, without compromising the quality of the model. Twenty-four male Sprague Dawley rats were injected with 20 μl of CFA into the left tibio-tarsal joint to induce monoarthritis. Rats were treated with either buprenorphine or carprofen for 15 days during the disease development, and were compared to a saline-treated CFA-injected group or a negative control group. Measurements of welfare, pain-related behaviour and clinical model-specific parameters were collected. The study was terminated after 3 weeks, ending with a histopathologic analysis. Regardless of treatment, CFA-injected rats displayed mechanical hyperalgesia and developed severe histopathological changes associated with arthritis. However, no severe effects on general welfare were found at any time. Buprenorphine treatment reduced facial pain expression scores, improved mobility, stance and lameness scores and it did not supress the CFA-induced ankle swelling, contrary to carprofen. Although buprenorphine failed to demonstrate a robust analgesic effect on the mechanical hyperalgesia in this study, it did not interfere with the development of the intended pathology.

Correction: Candidemia Risk Prediction (CanDETEC) Model for Patients With Malignancy: Model Development and Validation in a Single-Center Retrospective Study (Preprint)

UNSTRUCTURED REMOVE

Predicting relapse or recurrence of depression: systematic review of prognostic models

Background Relapse and recurrence of depression are common, contributing to the overall burden of depression globally. Accurate prediction of relapse or recurrence while patients are well would allow the identification of high-risk individuals and may effectively guide the allocation of interventions to prevent relapse and recurrence. Aims To review prognostic models developed to predict the risk of relapse, recurrence, sustained remission, or recovery in adults with remitted major depressive disorder. Method We searched the Cochrane Library (current issue); Ovid MEDLINE (1946 onwards); Ovid Embase (1980 onwards); Ovid PsycINFO (1806 onwards); and Web of Science (1900 onwards) up to May 2021. We included development and external validation studies of multivariable prognostic models. We assessed risk of bias of included studies using the Prediction model risk of bias assessment tool (PROBAST). Results We identified 12 eligible prognostic model studies (11 unique prognostic models): 8 model development-only studies, 3 model development and external validation studies and 1 external validation-only study. Multiple estimates of performance measures were not available and meta-analysis was therefore not necessary. Eleven out of the 12 included studies were assessed as being at high overall risk of bias and none examined clinical utility. Conclusions Due to high risk of bias of the included studies, poor predictive performance and limited external validation of the models identified, presently available clinical prediction models for relapse and recurrence of depression are not yet sufficiently developed for deploying in clinical settings. There is a need for improved prognosis research in this clinical area and future studies should conform to best practice methodological and reporting guidelines.

Gel Textural Characteristics of Hair Gel with Cocoa Shell Extract by Using Mixture D-optimal Method

This study was used a mixture design to optimize the spreadability and viscosity of topical hair gel incorporates cocoa shell extract. The factor of the hair gel ingredient was thickener (0.2 – 0.8%), styling polymer A (2-5%), styling polymer B (2-6%), and solvent (84.63-91.63%) were studied on two responses selected spreadability and viscosity. The data collected were fitted to the model with high coefficient determination (R2= 0.994 for the spreadability and 0.9937 for the viscosity). The model can be predicted by showing the good lack of fit test result not significant with the p-value bigger than 0.05. From the ramp function simulation, the optimized formulation was selected and established at thickener (0.55%), styling polymer A (3.61%), styling polymer B (3.72%), and solvent (88.55%) with the spreadability and viscosity at 353.77 g.s and 39.91 pa.s respectively. The benefit of using mixture design in this experiment, it can help a formulator to understand the complex interaction between factors and can easily modify the formulation through ramp function simulation to obtain the desired result. The predicted validation test shows that both values were comparable. Under this condition showed that the model development could be used to predict future observations within the design range thickener (0.2 – 0.8%), styling polymer A (2-5%), styling polymer B (2-6%), and solvent (84.63-91.63%).