Prediction Score Research Articles

A new signature associated with anoikis predicts the outcome and immune infiltration in nasopharyngeal carcinoma.

Previous studies have confirmed the phenomenon of anoikis resistance in nasopharyngeal carcinoma (NPC). Nevertheless, the prognostic significance of anoikis-related genes (ARGs) in NPC remains incompletely understood. This study aimed to create a predictive risk score using an ARGs signature for NPC patients and to investigate how this score relates to clinicopathologic features and immune infiltration in the tumor microenvironment. By using data from the Gene Expression Omnibus (GEO) database, we employed machine learning methods to discover prognostic ARGs and create a risk score. Key gene expression levels were validated through real-time PCR and immunohistochemical staining. Three differentially expressed ARGs (CDC25C, E2F1 and RBL2) with prognostic value were identified by the intersection of multiple machine learning algorithms. A risk score based on t 3-ARG feature was developed to stratify NPC patients into two distinct risk groups using the optimal model, Random Survival Forest. NPC patients with high-risk scores experienced notably shorter progression-free survival in comparison to those with low-risk scores. Multivariate Cox regression analysis indicated that the risk score served as an independent prognostic factor. The time-dependent ROC and decision curve analyses demonstrated the risk model's strong predictive accuracy and clinical utility. The low-risk score group exhibited features indicative of early clinical stage, immune activation, high immune checkpoint gene's expression, and low Epstein-Barr virus gene's expression. Functional analysis revealed enrichment of immune-related pathways in the low-risk group. Patients with high-risk scores were discovered to be unlikely to benefit from immune checkpoint inhibitor treatment. Moreover, the expression of RBL2, E2F1, and CDC25C were significantly correlated with the expression of caspase family genes. Finally, the lower mRNA and protein expression of RBL2 were validated in NPC cell lines and tissues. Our ARGs-based signature model shows promising results in predicting the prognosis of NPC patients and might be associated with immune cell infiltration.

Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI.

Gadolinium-enhanced "sampling perfection with application-optimized contrasts using different flip angle evolution" (SPACE) sequence allows better visualization of brain metastases (BMs) compared to "magnetization-prepared rapid acquisition gradient echo" (MPRAGE). We hypothesize that this better conspicuity leads to high-quality annotation (HAQ), enhancing deep learning (DL) algorithm detection of BMs on MPRAGE images. Retrospective contrast-enhanced (gadobutrol 0.1 mmol/kg) SPACE and MPRAGE data of 157 patients with BM were used, either annotated on MPRAGE resulting in normal annotation quality (NAQ) or on coregistered SPACE resulting in HAQ. Multiple DL methods were developed with NAQ or HAQ using either SPACE or MRPAGE images and evaluated on their detection performance using positive predictive value (PPV), sensitivity, and F1 score and on their delineation performance using volumetric Dice similarity coefficient, PPV, and sensitivity on one internal and four additional test datasets (660 patients). The SPACE-HAQ model reached 0.978 PPV, 0.882 sensitivity, and 0.916 F1-score. The MPRAGE-HAQ reached 0.867, 0.839, and 0.840, the MPRAGE NAQ 0.964, 0.667, and 0.798, respectively (p ≥ 0.157). Relative to MPRAGE-NAQ, the MPRAGE-HAQ F1-score detection increased on all additional test datasets by 2.5-9.6 points (p < 0.016) and sensitivity improved on three datasets by 4.6-8.5 points (p < 0.001). Moreover, volumetric instance sensitivity improved by 3.6-7.6 points (p < 0.001). HAQ improves DL methods without specialized imaging during application time. HAQ alone achieves about 40% of the performance improvements seen with SPACE images as input, allowing for fast and accurate, fully automated detection of small (< 1 cm) BMs. Training with higher-quality annotations, created using the SPACE sequence, improves the detection and delineation sensitivity of DL methods for the detection of brain metastases (BMs)on MPRAGE images. This MRI cross-technique transfer learning is a promising way to increase diagnostic performance. Delineating small BMs on SPACE MRI sequence results in higher quality annotations than on MPRAGE sequence due to enhanced conspicuity. Leveraging cross-technique ground truth annotations during training improved the accuracy of DL models in detecting and segmenting BMs. Cross-technique annotation may enhance DL models by integrating benefits from specialized, time-intensive MRI sequences while not relying on them. Further validation in prospective studies is needed.

Predictive score for response to neoadjuvant chemotherapy in early-stage HR + /HER2- breast cancer.

Neoadjuvant chemotherapy (NACT) is a treatment option for early-stage hormone receptor-positive human epidermal growth factor receptor 2-negative (HR + /HER2-) breast cancer. Despite its use, pathological response rates in this subtype are often lower, and the impact of individual risk factors remains unclear. This study aimed to identify biomarkers and create a predictive score for NACT response. This retrospective, single-center study included patients with stage IIA-IIIC HR + /HER2- breast cancer treated with NACT and surgery (2019-2023). Multiple logistic regression analyzed associations between clinicopathological variables and pathologic response (partial/complete vs. absent) (p < 0.05). The best-performing model was used to develop a risk score. The study included 101 patients. Significant predictors of pathological response included tumor grade (G2/3 vs. G1), menopausal status (pre- vs. post-menopausal), and intrinsic subtype (luminal B vs. A). A dynamic calculator was created incorporating grade, hormonal status, intrinsic subtype, and Ki-67. This tool provides real-time input, facilitating personalized therapeutic decision-making.

The external validity of machine learning-based prediction scores from hematological parameters of COVID-19: A study using hospital records from Brazil, Italy, and Western Europe.

The unprecedented worldwide pandemic caused by COVID-19 has motivated several research groups to develop machine-learning based approaches that aim to automate the diagnosis or screening of COVID-19, in large-scale. The gold standard for COVID-19 detection, quantitative-Real-Time-Polymerase-Chain-Reaction (qRT-PCR), is expensive and time-consuming. Alternatively, haematology-based detections were fast and near-accurate, although those were less explored. The external-validity of the haematology-based COVID-19-predictions on diverse populations are yet to be fully investigated. Here we report external-validity of machine learning-based prediction scores from haematological parameters recorded in different hospitals of Brazil, Italy, and Western Europe (raw sample size, 195554). The XGBoost classifier performed consistently better (out of seven ML classifiers) on all the datasets. The working models include a set of either four or fourteen haematological parameters. The internal performances of the XGBoost models (AUC scores range from 84% to 97%) were superior to ML models reported in the literature for some of these datasets (AUC scores range from 84% to 87%). The meta-validation on the external performances revealed the reliability of the performance (AUC score 86%) along with good accuracy of the probabilistic prediction (Brier score 14%), particularly when the model was trained and tested on fourteen haematological parameters from the same country (Brazil). The external performance was reduced when the model was trained on datasets from Italy and tested on Brazil (AUC score 69%) and Western Europe (AUC score 65%); presumably affected by factors, like, ethnicity, phenotype, immunity, reference ranges, across the populations. The state-of-the-art in the present study is the development of a COVID-19 prediction tool that is reliable and parsimonious, using a fewer number of hematological features, in comparison to the earlier study with meta-validation, based on sufficient sample size (n = 195554). Thus, current models can be applied at other demographic locations, preferably, with prior training of the model on the same population. Availability: https://covipred.bits-hyderabad.ac.in/home; https://github.com/debashreebanerjee/CoviPred.

Predicting the risk of invasive fungal infections in ICU sepsis population: the AMI risk assessment tool.

Invasive fungal infections (IFI) represent a significant contributor to mortality among sepsis patients in the Intensive Care Unit (ICU). Early diagnosis of IFI is challenging, and currently, there are no predictive tools for identifying sepsis patients who may develop IFI. Our study aims to develop a predictive scoring system to assess the risk of IFI in patients with sepsis admitted to the ICU. A retrospective collection of data from a total of 549 patients was conducted. Data-driven, clinically knowledge-driven, and decision tree models were used to identify predictive variables for risk of IFI in ICU patients with sepsis. Demographic data, vital signs, laboratory values, comorbidities, medication use, and clinical outcomes were all collected. The optimal model was selected based on model performance and clinical utility to establish a risk score. Among adult patients with sepsis admitted to the ICU, 127 patients (23.1%) developed IFI. The final data-driven model included four predictive factors, the clinically knowledge-driven model included three predictive factors, and the decision tree model included two. Based on the good performance and clinical utility of the clinically knowledge-driven model, it was chosen as the optimal risk scoring model (C-statistics: 0.79 (95% confidence interval (CI): 0.75-0.83); Hosmer-Lemeshow (H-L) test P = 0.884). The ICU sepsis patient invasive fungal infection risk (AMI) score, created based on the clinically knowledge-driven model, includes mechanical ventilation, application of immunosuppressants, and the types of antibiotics used. The C-statistics for this risk score was 0.79 (95% CI:0.75-0.84) with good calibration (H-L test P = 0.992 and see calibration curve: Fig.2). Moreover, in terms of clinical utility, the decision curve analysis for AMI showed a favorable net benefit. The application of the AMI score can effectively distinguish whether ICU sepsis patients will develop IFI, which is beneficial for clinicians to formulate targeted and timely preventive and treatment measures based on the risk of IFI.

Effectiveness of liver transplant mortality scales in a Mexican population.

Liver transplantation is the treatment of choice in patients with chronic liver disease and acute liver failure of any etiology. Scales such as the Survival Outcome Following Transplantation (SOFT) score and the Balance of Risk (BAR) score can be used to predict survival. In this study, we compared these scales in the Mexican population. A cross-sectional analytical study was carried out in a Mexican third-level transplant center. The MELD, SOFT, and BAR scales were adopted. The ROC curves of the three predictive scores were constructed, and the areas under the curve were obtained and compared. A bivariate analysis and Cox regression were performed. Finally, a survival analysis was performed using Kaplan-Meier curves. We analyzed 123 liver transplant (LT) recipients. The bivariate analysis and Cox regression indicated that portal thrombosis, with an HR of 3.36 (IC 1.069-10.59, p = 0.038), and the number of red blood cells transfused, with an HR of 1.084 (CI 1.039-1.130, p < 0.000), were significantly associated with mortality. The receiver height was a protective factor, with an HR of 0.001 (CI 0.000-0.761, p = 0.041). Regarding the Pearson correlation analysis, the BAR scale had a coefficient of 0.199 (p = 0.032) for transfusion, while the SOFT scale's correlation coefficients for cold ischemia and transfusion were 0.236 (p = 0.011) and 0.274 (p = 0.003), respectively, all indicating weak correlations. The areas under the curve (AUCs) of MELD, SOFT, and BAR in predicting 3-month mortality were 0.495 (P = 0.94), 0.608 (p = 0.129), and 0.502 (p = 0.97), respectively. Finally, in the survival analysis using Kaplan-Meier curves, an estimated mean survival period of 71.52 months was obtained, with a survival rate of 89.3 % at 30 days and 81.1 % at five years. In this study, it was found that all three scales were deficient in discriminating among the outcomes obtained in the Mexican population.

Preoperative prediction of flexible ureteroscopy outcome in the treatment of renal calculi

BackgroundRetrograde intrarenal surgery (RIRS) is increasingly used for management of renal calculi. Many predictive scores have been devised to predict stone-free rate (SFR) after RIRS including Ito’s nomogram. The aim of this study is to evaluate preoperative parameters that affect SFR after RIRS to help with patient counselling.MethodsPatients with renal stones up to 30 mm in greatest diameter were recruited. They were offered RIRS via disposable flexible ureteroscopy and Holmium laser fragmentation. Ito’s score was calculated for each patient preoperatively to estimate post-operative SFR. Also, other factors as stone Hounsfield units (HU), pre-stenting, and access sheath use were recorded.ResultsA total of 80 patients met the inclusion criteria. Median stone volume was smaller in cases that were stone free (532.4 vs 748.0 mm3, p = 0.076). Patients with residuals had higher rates of multiple (61.5% vs 16.7%, p < 0.001) and lower pole (61.5%vs 37.0%, p = 0.039) calculi. Hydronephrosis, operator experience, HU, access sheath use, pre-stenting, and total Ito score were not significantly different between stone-free patients and patients with residuals. Multivariate logistic regression showed that presence of multiple renal stones (B = 0.158, p = 0.002) is the only significant predictor of SFR.ConclusionLarger stones, stones in the lower pole, and especially multiple renal stones have higher rates of residual stones after RIRS. This should be clarified to patients during preoperative decision-making and discussion about RIRS and its alternatives’ pros and cons has to take place prior to surgery.Trial registration: Trial registered at clinicaltrial.gov, under number NCT05857501 on May 10th, 2023. https://clinicaltrials.gov/study/NCT05857501.

Investigating the frequency of neural autoantibodies in refractory focal epilepsy.

There have been conflicting reports about the frequency of neural autoantibodies in epilepsy cohorts, which is confounded by the lack of clear distinction of epilepsy from acute symptomatic seizures due to encephalitis. The aim of this study was to determine the frequency of neural autoantibodies in a well characterised population of refractory focal epilepsy of known and unknown cause. Cases were recruited from epilepsy outpatient clinics at the Princess Alexandra, Mater, Royal Brisbane and Women's and Cairns Base Hospitals from 2021 - 2023. Included cases were refractory to medical therapy, met the ILAE definition of focal epilepsy and were characterised using anatomo-electro-clinical correlation. Cases with prior encephalitis, inflammatory neurological disease or prior parenchymal brain insults were excluded. A total of 100 patients were recruited. No cases with clinically significant neural autoantibodies were discovered. One was positive for serum anti-NMDAR antibodies, however autoantibodies were absent from CSF. Cases were also screened using a predictive score (McGinty et al.). From this, 2 cases were identified as seronegative autoimmune associated epilepsy and demonstrated significant reduction in seizure frequency with administration of immunotherapy. These cases had common features including temporo-perisylvian semiology especially ictal piloerection and high seizure frequency. Clinically relevant neural autoantibodies are uncommon in well characterised chronic focal epilepsy populations. Despite this there are isolated cases that still demonstrate improved seizure control with the use of immunotherapy. Such cases highlight the need for further studies to understand the role of immunity in novel pathophysiological mechanisms in epilepsy.

Development and validation of a predictive score for diagnosing prostate cancer in primary care.

Prostate cancer (PCa) represents the fifth cause of death in the male population worldwide. The prostate-specific antigen (PSA) test demonstrated poor accuracy to assess the presence of PCa. Thus, the PSA testing paradigm should be moved from the systematic screening approach to the early identification of men who are harbouring clinically significant disease. Accurate clinical-based tools to predict PCa should therefore be developed for general practice. We derived and validated a PCa predictive score using a primary care data source. Using the Italian Health Search Database, we formed a cohort of men aged 45-90 years in the period between 2002 and 2015. These patients were followed up until 31 December 2022. Those with less than a 5-year follow-up were excluded. The cohort was randomly divided into 'derivation' and 'validation' samples in a 1:1 ratio. Along with the demographic and clinical determinants forming the score, we investigated the role of PSA kinetics in the prediction accuracy. In a cohort of 529,082 men aged 45+ years, we identified 14,524 cases of PCa (incidence rate = 2.71 per 1000 person-years; 95% confidence interval = 2.67-2.80). The prediction accuracy of the PCa-HScore featured an explained variation of 12% and a discrimination power of 70%. The calibration slope was almost equal to 1 (p = 0.951, tested for equivalence against the 'perfect' slope) and the PSA kinetics did not improve the prediction accuracy. The PCa-HScore might guide the prescription of PSA and/or other clinical strategies in those men reporting certain levels of risk. A related decision support system could therefore be implemented in primary care.

Development and validation of a model to predict cognitive impairment in traumatic brain injury patients: a prospective observational study.

Traumatic brain injury (TBI) is a significant public health issue worldwide that affects millions of people every year. Cognitive impairment is one of the most common long-term consequences of TBI, seriously affect the quality of life. We aimed to develop and validate a predictive model for cognitive impairment in TBI patients, with the goal of early identification and support for those at risk of developing cognitive impairment at the time of hospital admission. The training cohort included 234 TBI patients, all of whom were admitted to the Department of Neurosurgery at the Third Affiliated Hospital of Soochow University from May 2017 to April 2020. These patients were selected from our previously published studies. Baseline characteristics, medical history, clinical TBI characteristics, treatment details, and vital signs during hospitalization were screened via least absolute shrinkage and selection operator (LASSO) and logistic regression to construct a predictive net risk score. The derived score represents an estimate of the risk of developing cognitive impairment in patients with TBI. A nomogram was constructed, and its accuracy and predictive performance were evaluated with the area under the receiver operating characteristic curve (AUC), calibration curves, and clinical decision curves. For the validation cohort, data were prospectively collected from TBI patients admitted to the Department of Neurosurgery at the Third Affiliated Hospital of Soochow University from March 1, 2024 to August 30, 2024, according to the inclusion and exclusion criteria. This study is registered with the Chinese Clinical Trial Registry (ChiCTR) at http://www.chictr.org.cn/ (registration number: ChiCTR2400083495). The training cohort included 234 patients. The mean (standard deviation, SD) age of the patients in the cohort was 47.74 (17.89) years, and 184 patients (78.63%) were men. The validation cohort included 84 patients with a mean (SD) age of 48.44 (14.42) years, and 68 patients (80.95%) were men. Among the 48 potential predictors, the following 6 variables were significant independent predictive factors and were included in the net risk score: age (odds ratio (OR)=1.06, 95% confidence interval (CI): 1.03-1.08, P=0.00), years of education (OR=0.80, 95% CI: 0.70-0.93, P=0.00), pulmonary infection status (OR=4.64, 95% CI: 1.41-15.27, P=0.01), epilepsy status (OR=4.79, 95% CI: 1.09-21.13, P=0.04), cerebrospinal fluid leakage status (OR=5.57, 95% CI: 1.08-28.75, P=0.04), and the Helsinki score (OR=1.53, 95% CI: 1.28-1.83, P=0.00). The AUC in the training cohort was 0.90, and the cut-off value was 0.71. The AUC in the validation cohort was 0.87, and the cut-off value was 0.63. The score was translated into an online risk calculator that is freely available to the public (https://yuanxiaofang.shinyapps.io/Predict_cognitive_impairment_in_TBI/). This model for predicting post-TBI cognitive impairment has potential value for facilitating early predictions by clinicians, aiding the early initiation of preventative interventions for cognitive impairment. This research was supported by Science and Technology Development Plan Project of ChangZhou (CJ20229036); Science and Technology Project of Changzhou Health Commission (QN202113).

Abstract WMP39: First External Validation of the Predictive Swallow Score

Abstract WMP39: First External Validation of the Predictive Swallow Score

Assessing hazard prediction and risk calibration skills in experienced and novice e-scooter riders

Less experienced e-scooter riders often exhibit risky riding behaviours. Despite this, no studies have examined how riders calibrate risk, respond to hazardous situations, and the impact of riding experience on these skills. To address this, this study assessed hazard prediction and risk calibration in e-scooter riders via bespoke video-based tests featuring real e-scooter footage filmed from the rider’s perspective. The first experiment assessed the ability of e-scooter riders to predict hazardous riding scenarios. The second experiment evaluated their proneness to engage in risky riding situations. The results indicated that increased riding experience did not improve riders’ hazard prediction skills or reduced their proneness to engage in risky riding. In fact, a higher riding frequency was linked to an increased tendency to engage in risky behaviour in certain scenarios. The results highlight that the typically short duration of e-scooter trips may limit riders’ exposure to a variety of hazards, hindering their ability to develop effective risk calibration skills. The observed high propensity to engage in risky riding scenarios, combined with average hazard prediction scores, emphasizes the need for targeted rider training focused on vigilance and risk awareness.

Development and validation of an automated predictive scoring system to identify patients at increased risk for Fabry disease using Japanese electronic cardiac failure data

Development and validation of an automated predictive scoring system to identify patients at increased risk for Fabry disease using Japanese electronic cardiac failure data

Predicting good clinical outcomes after revision rotator cuff repair: the revision cuff repair prediction score

Predicting good clinical outcomes after revision rotator cuff repair: the revision cuff repair prediction score

Dual versus single immune checkpoint blockade for patients with metastatic colorectal cancer harboring microsatellite instability: The Saint-Antoine immunoMSI score.

168 Background: Anti-PD1 monotherapy (mono) and anti-PD1 plus anti-CTLA4 therapies (combo) are validated for metastatic colorectal cancer (mCRC) harboring microsatellite instability and/or mismatch repair deficiency (MSI/dMMR). We aimed at assessing the efficacy of combined anti-CTLA4 and anti-PD1 blockade versus anti-PD1 monotherapy and identifying which subgroups benefit more from the combo. Methods: All patients (pts) with MSI/dMMR mCRC treated by anti-PD1 ± anti-CTLA4 included in the prospective monocenter immunoMSI cohort were analyzed. The choice between monotherapy or combination therapy varied according to open studies and regulatory approvals, without taking pt characteristics into account. Datalock date was April 23, 2024. Main endpoints were progression-free survival (PFS) per iRECIST and overall survival (OS) estimated with the Kaplan-Meier method. Univariate Cox proportional hazards models were used to estimate hazard ratios with CI. The interactions between studied subgroups and the treatment group were modelized by adding in a single Cox model the studied subgroups, the treatment group parameters, and their interaction term. Results: Among 210 pts, 97 were treated combo and 113 with mono. Baseline characteristics were similar in both groups. 25% and 17% were in first line, respectively. Median follow-up was 4.4 years: 4.1 years for combo and 5.4 years for mono. Combo was associated with better PFS (HR=0.48; 95%CI 0.31-0.76) and OS (HR=0.49; 0.29-0.84). Best observed response rates were: 78% of objective responses, 17% of stable diseases and 5% disease progressions with combo versus 60%, 25% and 15% with mono. Subgroup analyses showed significant differential effect of the treatment group for gender (PFS HR=0.21 vs 0.87 for females and males, P-interaction: 0.0067), liver metastases (PFS HR=0.30 vs 0.79 for pts without liver mets and those with liver mets, P-interaction: 0.0479) and sidedness (PFS HR=0.27 vs 0.64 for left-sided and right-sided mCRC, P-interaction=0.0922), but not RAS / RAF mutations. We computed a predictive score based on gender (female: 2 points), no liver met (1 point) and left side (1 point). PFS HR was 0.1 (0.03-0.33) for the group ≥3 points (N=67), 0.66 (0.36-1.23) for the group with 1-2 points (N=113) and 1.54 (0.50-4.70) for the group with 0 point (N=30). Conclusions: Combined CTLA4 and PD1 blockade is associated with improved PFS and OS compared to anti-PD1 alone in MSI/dMMR mCRC. The subgroup deriving the highest benefit of the combo seems to be female pts with left sided tumor and no liver met. Data from the randomization between nivolumab versus nivolumab + ipilumumab in the CheckMate-8HW phase III study are expected.

Scoring Strategic Agents

I introduce a model of predictive scoring. A receiver wants to predict a sender’s quality. An intermediary observes multiple features of the sender and aggregates them into a score. Based on the score, the receiver makes a decision. The sender prefers “higher” decisions, and she can distort each feature at a privately known cost. I characterize the scoring rule that maximizes decision accuracy. This rule underweights some features to deter sender distortion, and overweights other features so that the score is correct on average. The receiver prefers this scoring rule to full disclosure because it mitigates his commitment problem. (JEL C72, D82, D83)

The inflation-uncertainty nexus: new evidence from stochastic volatility models

ABSTRACT The purpose of this study is to examine the inflation and inflation uncertainty nexus for nine emerging and three developed market economies using more realistic and flexible inflation uncertainty measures based on seven different stochastic volatility (SV) models. In that regard, we first utilize a log predictive score to determine the most appropriate SV model for each economy examined, and then analyse the causality relationship between inflation and inflation uncertainty. Contrary to general findings in the related literature, the results provide strong and robust evidence that inflation uncertainty has a positive and statistically significant impact on the inflation rates for both the emerging and developed countries examined.

Fully automated coronary artery calcium score and risk categorization from chest CT using deep learning and multiorgan segmentation: A validation study from National Lung Screening Trial (NLST).

The National Lung Screening Trial (NLST) has shown that screening with low dose CT in high-risk population was associated with reduction in lung cancer mortality. These patients are also at high risk of coronary artery disease, and we used deep learning model to automatically detect, quantify and perform risk categorisation of coronary artery calcification score (CACS) from non-ECG gated Chest CT scans. Automated calcium quantification was performed using a neural network based on Mask regions with convolutional neural networks (R-CNN) for multiorgan segmentation. Manual evaluation of calcium was carried out using proprietary software. This study used 80 patients to train the segmentation model and randomly selected 1442 patients were used for the validation of the algorithm. We compared the model generated results with Ground Truth. Automatic cardiac and aortic segmentation model worked well (Mean Dice score: 0.91). Cohen's kappa coefficient between the reference actual and the interclass computed predictive categories on the test set is 0.72 (95% CI: 0.61-0.83). Our method correctly classifies the risk group in 78.8% of the cases and classifies the subjects in the same group. F-score is measured as 0.78; 0.71; 0.81; 0.82; 0.92 in calcium score categories 0(CS:0), I (1-99), II (100-400), III (400-1000), IV (>1000), respectively. 79% of the predictive scores lie in the same categories, 20% of the predictive scores are one category up or down, and only 1.2% patients were more than one category off. For the presence/absence of coronary artery calcifications, our deep learning model achieved a sensitivity of 90% and a specificity of 94%. Fully automated model shows good correlation compared with reference standards. Automating the process could improve diagnostic ability, risk categorization, facilitate primary prevention intervention, improve morbidity and mortality, and decrease healthcare costs.

Use of SAPS 3, APACHE IV, and GRACE as prognostic scores for acute coronary syndrome patients in the cardiovascular care unit

A grading system based on disease severity has been widely used in intensive care units (ICUs) since around 1980. These systems are used to predict mortality and assess severity in clinical trials. Simplified Acute Physiology Score 3 (SAPS3) and Acute Physiology and Chronic Health Evaluation Score (APACHE IV) are prognosis ratings that can predict in-hospital mortality within the first hour of ICU care. Although these technologies have been widely employed in the ICU, they have yet to be commonly deployed in the cardiovascular care unit (CVCU) due to different patient populations. Intensive care doctors typically employ the standard prognostic scores, SAPS3 and APACHE IV, which were generated from diverse populations of critically ill patients. Although these scores are the most widely used early versions, APACHE IV and SAPS 3 do not include acute coronary syndrome patients. The Global Registry of Acute Coronary Events (GRACE) score has performed the best; this may be because of its straightforward design, which does not distinguish between individuals with SCA and those without ST-segment elevation. Our review article attempts to evaluate the performance of standard predictor scores, namely SAPS 3, APACHE IV, and GRACE, on patients with cardiovascular emergencies. Thus, these score systems can precisely assess the relationship between mortality prediction scores and outcomes of patients admitted to the CVCU rapidly and comprehensively. Keywords: Simplified Acute Physiology Score 3, Acute Physiology and Chronic Health Evaluation Score, Global Registry of Acute Coronary Events, intensive cardiovascular care unit

Predictors of Perioperative Outcomes in Chronic Limb-threatening Ischemia Patients

Background: Revascularization in chronic limb-threatening ischemia (CLTI) patients carries a risk of perioperative events. Scoring systems such as the Society for Vascular Surgery vascular quality initiative mortality prediction score (VQI MPS) for 30 days and 2-year mortality and revised cardiac risk index (RCRI) for perioperative major adverse cardiac events (MACE) have been validated to risk stratify patients. Both scoring systems had not been previously validated in an Indian cohort. Aims and Objectives: This study was conducted to validate both scoring systems and to identify predictors of perioperative adverse events. The primary objective was the validation of scoring systems based on observed peri-operative 30-day mortality and MACE. The secondary objective was to identify risk factors that affected these outcomes. Materials and Methods: Data of subsequent CLTI patients who underwent open or endovascular revascularization in a single centre between 2018 and 2024 were retrieved from a prospectively maintained database and retrospectively analysed. Patients who were lost to follow-up or had incomplete data were excluded from the study. MACE was defined as any acute coronary syndrome, pulmonary oedema, or cardiac-related death. Validity of VQI MPS and RCRI was calculated using receiver-operating curves. Predictors of outcome were analysed by univariate analysis and logistic regression. Results: One hundred forty-six patients were analysed and risk-stratified based on VQI MPS and RCRI. The mean age was 68 ± 9 years with 31.5% females. About 88.4% (n = 129/146), 9.6% (n = 14/146), and 2% (n = 3/146) were categorized into low, medium, and high risk as per VQI MPS and 19 (13.05%), 59 (40.4%), 49 (33.5%), and 19 (13.05%) patients were categorized into Class 1, II, III, IV and above according to RCRI, respectively. Both scoring systems accurately risk-stratified procedures into different strata. Seven (4.79%) patients died within 30 day of procedure and 10 (6.84%) patients had MACE. There was a statistical difference between the predicted versus observed 30-day mortality and MACE. The discriminatory power of VQI MPS and RCRI were poor with c statistic of 0.664 for VQI MPS and 0.518 for RCRI. Univariate analysis showed that congestive heart failure (CHF) was the only factor that affected 30-day mortality and perioperative MACE. Conclusion: VQI MPS and RCRI did not accurately predict perioperative outcomes in CLTI patient undergoing revascularization in an Indian population. CHF was found to be the only predictor of perioperative mortality and MACE.