A practical solution to estimate the sample size required for clinical prediction models generated from observational research on data

Clinical prediction model for red cell blood transfusion in elective primary posterior lumbar spine fusion

Overestimated the cross-match of preoperative PRC preparation for elective primary lumbar spinal fusion needs revision for cost-effectiveness. We aimed to develop a novel preoperative predictive model for appropriate PRC preparation. This clinical prediction model in a retrospective cohort was studied between January 2015 and September 2022. Multivariate logistic regression models were used to assess predictive variables. The logistic coefficient of each predictor generated scores to establish a predictive model. The area under the receiver operating characteristic curve (AuROC) was used to evaluate the model. The predictive performance was validated using bootstrapping techniques and externally validated in 102 independent cases. Among 416 patients, 178 (43%) required transfusion. Four final predictors: preoperative hematocrit level, laminectomy level, transforaminal lumbar interbody fusion level, and sacral fusion. When categorized into two risk groups, the positive predictive values for the low-risk score (≤ 4) were 18.4 (95% Cl 13.9, 23.6) and 83.9 (95% CI 77.1, 89.3) for the high-risk score (> 4). AuROC was 0.90. Internal validation (bootstrap shrinkage = 0.993) and external validation (AuROC: 0.91). A new model demonstrated exemplary performance and discrimination in predicting the appropriate preparation for PRC. This study should be corroborated by rigorous external validation in other hospitals and by prospective assessments.

Development of a multivariate prediction model for antidepressant resistant depression using reward-related predictors

Introduction

Individuals with depression who do not respond to two or more courses of serotonergic antidepressants tend to have greater deficits in reward processing and greater internalizing symptoms, yet there is no validated self-report method to determine the likelihood of treatment resistance based on these symptom dimensions.

Methods

This online case-control study leverages machine learning techniques to identify differences in self-reported anhedonia and internalizing symptom profiles of antidepressant non-responders compared to responders and healthy controls, as an initial proof-of-concept for relating these indicators to medication responsiveness. Random forest classifiers were used to identify a subset from a set of 24 reward predictors that distinguished among serotonergic medication resistant, non-resistant, and non-depressed individuals recruited online (N = 393). Feature selection was implemented to refine model prediction and improve interpretability.

Results

Accuracies for full predictor models ranged from .54 to .71, while feature selected models retained 3-5 predictors and generated accuracies of .42 to .70. Several models performed significantly above chance. Sensitivity for non-responders was greatest after feature selection when compared to only responders, reaching .82 with 3 predictors. The predictors retained from feature selection were then explored using factor analysis at the item level and cluster analysis of the full data to determine empirically driven data structures.

Discussion

Non-responders displayed 3 distinct symptom profiles along internalizing dimensions of anxiety, anhedonia, motivation, and cognitive function. Results should be replicated in a prospective cohort sample for predictive validity; however, this study demonstrates validity for using a limited anhedonia and internalizing self-report instrument for distinguishing between antidepressant resistant and responsive depression profiles.

External validation of novel Revised Intensity Battle Score and comparison of static rib fracture scoring systems

BACKGROUND

This study aims to compare and externally validate the previously developed Revised Intensity Battle Score (RIBS) against other proposed scores for predicting poor outcomes after rib fractures.

METHODS

An external validation set was assembled retrospectively, comprising 1,493 adult patients with one or more rib fractures admitted to a Level 1 trauma center between 2019 and 2022. The following rib fracture scores were calculated for each patient: RIBS, Injury Severity Score, Rib Fracture Score, Chest Trauma Score, and Battle score. Each was investigated to assess utility in predicting mortality, intensive care unit upgrade, unplanned intubation and ventilator days. Performance was measured by area under the receiver operating characteristic curve.

RESULTS

Of the 1,493 patients who met inclusion criteria, 239 patients (16%) experienced one of more of the investigated outcomes. Generally, scores performed best at predicting mortality and ventilator days. The RIBS stood out as best predicting "any complication" (AUC = 0.735) and ">7 ventilator days" (AUC = 0.771).

CONCLUSION

The RIBS represents an externally validated triage score in patients with rib fractures and compares favorably to other static scoring systems. Use of this score as a triage tool may allow stratifying patients who may benefit from direct intensive care unit admission, neuraxial anesthesia and aggressive respiratory care. Next steps include prospective investigation of how pairing these interventions with score directed triage impacts outcomes.

LEVEL OF EVIDENCE

Prognostic and Epidemiological; Level IV.

The role of intimate partner violence perpetrators' resting state functional connectivity in treatment compliance and recidivism

To expand the scientific literature on how resting state functional connectivity (rsFC) magnetic resonance imaging (MRI) (or the measurement of the strength of the coactivation of two brain regions over a sustained period of time) can be used to explain treatment compliance and recidivism among intimate partner violence (IPV) perpetrators. Therefore, our first aim was to assess whether men convicted of IPV (n = 53) presented different rsFC patterns from a control group of non-violent (n = 47) men. We also analyzed if the rsFC of IPV perpetrators before staring the intervention program could explain treatment compliance and recidivism one year after the intervention ended. The rsFC was measured by applying a whole brain analysis during a resting period, which lasted 45 min. IPV perpetrators showed higher rsFC in the occipital brain areas compared to controls. Furthermore, there was a positive association between the occipital pole (OP) and temporal lobes (ITG) and a negative association between the occipital (e.g., occipital fusiform gyrus, visual network) and both the parietal lobe regions (e.g., supramarginal gyrus, parietal operculum cortex, lingual gyrus) and the putamen in IPV perpetrators. This pattern was the opposite in the control group. The positive association between many of these occipital regions and the parietal, frontal, and temporal regions explained treatment compliance. Conversely, treatment compliance was also explained by a reduced rsFC between the rostral prefrontal cortex and the frontal gyrus and both the occipital and temporal gyrus, and between the temporal and the occipital and cerebellum areas and the sensorimotor superior networks. Last, the enhanced rsFC between the occipital regions and both the cerebellum and temporal gyrus predicted recidivism. Our results highlight that there are specific rsFC patterns that can distinguish IPV perpetrators from controls. These rsFC patterns could be useful to explain treatment compliance and recidivism among IPV perpetrators.

A prognostic model and pre-discharge predictors of post-COVID-19 syndrome after hospitalization for SARS-CoV-2 infection

Background

Post-COVID-19 syndrome (PCS) has been increasingly recognized as an emerging problem: 50% of patients report ongoing symptoms 1 year after acute infection, with most typical manifestations (fatigue, dyspnea, psychiatric and neurological symptoms) having potentially debilitating effect. Early identification of high-risk candidates for PCS development would facilitate the optimal use of resources directed to rehabilitation of COVID-19 convalescents.

Objective

To study the in-hospital clinical characteristics of COVID-19 survivors presenting with self-reported PCS at 3 months and to identify the early predictors of its development.

Methods

221 hospitalized COVID-19 patients underwent symptoms assessment, 6-min walk test, and echocardiography pre-discharge and at 1 month; presence of PCS was assessed 3 months after discharge. Unsupervised machine learning was used to build a SANN-based binary classification model of PCS development.

Results

PCS at 3 months has been detected in 75% patients. Higher symptoms level in the PCS group was not associated with worse physical functional recovery or significant echocardiographic changes. Despite identification of a set of pre-discharge predictors, inclusion of parameters obtained at 1 month proved necessary to obtain a high accuracy model of PCS development, with inputs list including age, sex, in-hospital levels of CRP, eGFR and need for oxygen supplementation, and level of post-exertional symptoms at 1 month after discharge (fatigue and dyspnea in 6MWT and MRC Dyspnea score).

Conclusion

Hospitalized COVID-19 survivors at 3 months were characterized by 75% prevalence of PCS, the development of which could be predicted with an 89% accuracy using the derived neural network-based classification model.

A multiscale orchestrated computational framework to reveal emergent phenomena in neuroblastoma

Neuroblastoma is a complex and aggressive type of cancer that affects children. Current treatments involve a combination of surgery, chemotherapy, radiotherapy, and stem cell transplantation. However, treatment outcomes vary due to the heterogeneous nature of the disease. Computational models have been used to analyse data, simulate biological processes, and predict disease progression and treatment outcomes. While continuum cancer models capture the overall behaviour of tumours, and agent-based models represent the complex behaviour of individual cells, multiscale models represent interactions at different organisational levels, providing a more comprehensive understanding of the system. In 2018, the PRIMAGE consortium was formed to build a cloud-based decision support system for neuroblastoma, including a multi-scale model for patient-specific simulations of disease progression. In this work we have developed this multi-scale model that includes data such as patient's tumour geometry, cellularity, vascularization, genetics and type of chemotherapy treatment, and integrated it into an online platform that runs the simulations on a high-performance computation cluster using Onedata and Kubernetes technologies. This infrastructure will allow clinicians to optimise treatment regimens and reduce the number of costly and time-consuming clinical trials. This manuscript outlines the challenging framework's model architecture, data workflow, hypothesis, and resources employed in its development.

Is Metabolic Score for Visceral Fat (METS-VF) a Better Index Than Other Adiposity Indices for the Prediction of Visceral Adiposity

Background

Visceral adiposity is an important risk factor for cardiometabolic diseases.

Objective

To determine whether the Metabolic Score for Visceral Fat (METS-VF) is more effective than other adiposity indices in predicting visceral fat area (VFA).

Methods

In this single-center and cross-sectional study, we included patients aged 20-50 years, without diabetes and coronary artery disease, who underwent computed tomography (CT) including the third lumbar vertebra. Age, blood pressure, waist circumference (WC), hip circumference, fasting lipids, and glucose were assessed. VFA was measured by cross-sectional examination of CT. The correlation of WC, body mass index (BMI), waist-hip ratio (WHR), lipid accumulation product (LAP), visceral adiposity index (VAI), a body shape index (ABSI), body roundness index (BRI), and METS-VF with VFA was analyzed by correlation analysis. The cut-off values and area under the curve (AUC) for identifying increased VFA (>130 cm2) were determined.

Results

We included 185 individuals with mean age 38.2 ± 8 and female predominance (58.4%). There was a significant positive correlation between all indices and VFA (p<0.001). ROC analysis revealed that METS-VF and WC demonstrated the highest predictive value for identifying increased VFA. In both men (p=0.001) and women (p<0.001), METS-VF (AUC 0.922 and 0.939, respectively) showed a significant superiority over ABSI (AUC 0.702 and 0.658, respectively), and VAI (AUC 0.731 and 0.725, respectively). Additionally, in women, its superiority over WHR (AUC 0.807) was also statistically significant (p=0.003). We identified a METS-VF cut-off point >6.4 in males >6.5 in females and WC cut-off point >88 cm in males (AUC 0.922), >90.5 cm in females (AUC 0.938).

Conclusion

METS-VF is strongly associated with visceral adiposity and better to predict increased VFA. However, its superiority over WC, BMI, BRI, and LAP was not significant. The results emphasize that WC is more appealing as screening indicator for visceral adiposity considering its easy use.

Clinical Trial Registry Name

Clinicaltrials.gov (http://www.clinicaltrials.gov).

Clinical Trial Registry Url

https://clinicaltrials.gov/ct2/show/NCT05648409.

Clinical Trial Registry Number

NCT05648409.

Spontaneous physical functional recovery after hospitalization for COVID-19: insights from a 1 month follow-up and a model to predict poor trajectory

Background

Long COVID syndrome has emerged as a new global healthcare challenge, with impaired physical performance being a prominent debilitating factor. Cardiopulmonary rehabilitation is a mainstay of management of symptomatic post-COVID patients, and optimization of candidate selection might allow for more effective use of available resources.

Methods

In order to study the natural dynamics and to identify predictors of physical functional recovery following hospitalization for COVID-19, 6 min walk test was performed pre-discharge in 176 patients (40% hypertensive, 53% female, mean age 53.2 ± 13.5 years) with re-evaluation at 1 month.

Results

Six min walk distance and the reached percent of predicted distance (6MWD%) were suboptimal at both visits-396 ± 71 m (68.7 ± 12.4%) pre-discharge and 466 ± 65 m (81.8 ± 13.6%) at 1 month. Associated changes included significant oxygen desaturation (2.9 ± 2.5 and 2.3 ± 2.2%, respectively) and insufficient increment of heart rate during the test (24.9 ± 17.5 and 28.2 ± 12.0 bpm) that resulted in low reached percent of individual maximum heart rate (61.1 ± 8.1 and 64.3 ± 8.2%). Automatic clusterization of the study cohort by the 6MWD% changes has allowed to identify the subgroup of patients with poor "low base-low increment" trajectory of spontaneous post-discharge recovery that were characterized by younger age (38.2 ± 11.0 vs. 54.9 ± 12.1, p < 0.001) but more extensive pulmonary involvement by CT (43.7 ± 8.8 vs. 29.6 ± 19.4%, p = 0.029) and higher peak ESR values (36.5 ± 9.7 vs. 25.6 ± 12.8, p < 0.001). Predictors of poor recovery in multivariate logistic regression analysis included age, peak ESR, eGFR, percentage of pulmonary involvement by CT, need for in-hospital oxygen supplementation, SpO2 and mMRC dyspnea score pre-discharge, and history of hypertension.

Conclusion

COVID-19 survivors were characterized by decreased physical performance pre-discharge as assessed by the 6 min walk test and did not completely restore their functional status after 1 month of spontaneous recovery, with signs of altered blood oxygenation and dysautonomia contributing to the observed changes. Patients with poor "low base-low increment" trajectory of post-discharge recovery were characterized by younger age but more extensive pulmonary involvement and higher peak ESR values. Poor post-discharge recovery in the study cohort was predictable by the means of machine learning-based classification model that used age, history of hypertension, need for oxygen supplementation, and ESR as inputs.

MAIC-10 brief quality checklist for publications using artificial intelligence and medical images

The use of artificial intelligence (AI) with medical images to solve clinical problems is becoming increasingly common, and the development of new AI solutions is leading to more studies and publications using this computational technology. As a novel research area, the use of common standards to aid AI developers and reviewers as quality control criteria will improve the peer review process. Although some guidelines do exist, their heterogeneity and extension advocate that more explicit and simple schemes should be applied on the publication practice. Based on a review of existing AI guidelines, a proposal which collects, unifies, and simplifies the most relevant criteria was developed. The MAIC-10 (Must AI Criteria-10) checklist with 10 items was implemented as a guide to design studies and evaluate publications related to AI in the field of medical imaging. Articles published in Insights into Imaging in 2021 were selected to calculate their corresponding MAIC-10 quality score. The mean score was found to be 5.6 ± 1.6, with critical items present in most articles, such as "Clinical need", "Data annotation", "Robustness", and "Transparency" present in more than 80% of papers, while improvements in other areas were identified. MAIC-10 was also observed to achieve the highest intra-observer reproducibility when compared to other existing checklists, with an overall reduction in terms of checklist length and complexity. In summary, MAIC-10 represents a short and simple quality assessment tool which is objective, robust and widely applicable to AI studies in medical imaging.

Development and Validation of a Prognostic Model and Bedside Score for the Neurological Outcome in Children with Tuberculous Meningitis

Only a few studies have explored prognostic factors for tuberculous meningitis (TBM) in children, and an easily applicable bedside prognostic score for TBM has not been developed yet. We compared the sociodemographic, clinical, radiological, and cerebrospinal fluid parameters in the cohort of 94 TBM cases aged 1 to 18 years, with at least 6 months of completed follow-up and determined the prognostic factors associated with poor functional outcome. We assessed our proposed prognostic model using both discrimination and calibration and subsequently used the bootstrap method to validate the model internally. We finally derived an easily applicable bedside prognostic score by rounding off the regression coefficients to the nearest integers. A total of 39 (41%) and 55 (59%) patients had poor and good functional outcomes, respectively, at the end of 6 months (12 died, 13%). In multivariate analysis, a high baseline Pediatric Cerebral Performance Category (PCPC) score, brain infarction in neuroimaging, tonic motor posturing, younger age, and stage III TBM were independent predictors of poor functional outcomes. The final model showed good discrimination (area under the curve = 88.2%, P < 0.001) and good calibration (Hosmer-Lemeshow test, P = 0.53). Bootstrapping also confirmed the internal validity of this model. The PITAS (PCPC score [P], brain infarction in neuroimaging [I], tonic motor posturing [T], age [A], and stage of TBM [S]) score developed from this model has a score ranging from 0 to 12, with a higher score predicting a higher risk of poor functional outcome. The PITAS score performed better than medical research council staging alone in predicting poor functional outcomes (area under the curve = 87.1% versus 82.3%). Our study's PITAS score, developed and internally validated, has good sensitivity and specificity in predicting poor functional outcomes in pediatric TBM cases at 6 months.

Considerations for artificial intelligence clinical impact in oncologic imaging: an AI4HI position paper

To achieve clinical impact in daily oncological practice, emerging AI-based cancer imaging research needs to have clearly defined medical focus, AI methods, and outcomes to be estimated. AI-supported cancer imaging should predict major relevant clinical endpoints, aiming to extract associations and draw inferences in a fair, robust, and trustworthy way. AI-assisted solutions as medical devices, developed using multicenter heterogeneous datasets, should be targeted to have an impact on the clinical care pathway. When designing an AI-based research study in oncologic imaging, ensuring clinical impact in AI solutions requires careful consideration of key aspects, including target population selection, sample size definition, standards, and common data elements utilization, balanced dataset splitting, appropriate validation methodology, adequate ground truth, and careful selection of clinical endpoints. Endpoints may be pathology hallmarks, disease behavior, treatment response, or patient prognosis. Ensuring ethical, safety, and privacy considerations are also mandatory before clinical validation is performed. The Artificial Intelligence for Health Imaging (AI4HI) Clinical Working Group has discussed and present in this paper some indicative Machine Learning (ML) enabled decision-support solutions currently under research in the AI4HI projects, as well as the main considerations and requirements that AI solutions should have from a clinical perspective, which can be adopted into clinical practice. If effectively designed, implemented, and validated, cancer imaging AI-supported tools will have the potential to revolutionize the field of precision medicine in oncology.