Clinical Practice For Prediction Research Articles

Prediction of prognosis in patients with panic disorder using pre-treatment brain white matter features

BackgroundThe early identification of patients with panic disorder (PD) with a poor prognosis is important for improving treatment outcomes; however, it is challenging due to a lack of objective biomarkers. We investigated the reliability of characterizing structural white matter (WM) connectivity and its ability to predict PD prognosis after pharmacotherapy. MethodsA total of 138 patients (59 men) with PD and 153 healthy controls (HCs; 73 men) participated in this study. PD symptom severity was measured using the Panic Disorder Severity Scale (PDSS) at baseline and follow-up periods of 8 weeks, 6 months, and 1 year. The least absolute shrinkage and selection operator (Lasso) was utilized to identify prognosis-related WM regions on diffusion imaging features. ResultsLasso identified seven prognosis-related WM regions: the bilateral posterior corona radiata, bilateral posterior limb of the internal capsule, the left retrolenticular part of the internal capsule, the left sagittal stratum, and the right fornix/stria terminalis. Some of these regions showed lower mean fractional anisotropy (FA) values in patients with PD than in HCs. The predicted PDSS scores using FA from these regions consistently correlated with the actual prognosis in all periods. LimitationsThis study had limited ability to evaluate individual longitudinal changes in detail owing to the data acquisition time and brain atlas resolution. ConclusionsOur findings suggest the possibility of using structural WM connectivity as a biomarker for the clinical characterization of PD. Our findings will expand our understanding of the neurobiology of PD and improve biomarker-based prognosis prediction in clinical practice.

Machine learning model to predict mental health crises from electronic health records

The timely identification of patients who are at risk of a mental health crisis can lead to improved outcomes and to the mitigation of burdens and costs. However, the high prevalence of mental health problems means that the manual review of complex patient records to make proactive care decisions is not feasible in practice. Therefore, we developed a machine learning model that uses electronic health records to continuously monitor patients for risk of a mental health crisis over a period of 28 days. The model achieves an area under the receiver operating characteristic curve of 0.797 and an area under the precision-recall curve of 0.159, predicting crises with a sensitivity of 58% at a specificity of 85%. A follow-up 6-month prospective study evaluated our algorithm’s use in clinical practice and observed predictions to be clinically valuable in terms of either managing caseloads or mitigating the risk of crisis in 64% of cases. To our knowledge, this study is the first to continuously predict the risk of a wide range of mental health crises and to explore the added value of such predictions in clinical practice.

ICU Cockpit: a platform for collecting multimodal waveform data, AI-based computational disease modeling and real-time decision support in the intensive care unit.

ICU Cockpit: a secure, fast, and scalable platform for collecting multimodal waveform data, online and historical data visualization, and online validation of algorithms in the intensive care unit. We present a network of software services that continuously stream waveforms from ICU beds to databases and a web-based user interface. Machine learning algorithms process the data streams and send outputs to the user interface. The architecture and capabilities of the platform are described. Since 2016, the platform has processed over 89 billion data points (N = 979 patients) from 200 signals (0.5-500Hz) and laboratory analyses (once a day). We present an infrastructure-based framework for deploying and validating algorithms for critical care. The ICU Cockpit is a Big Data platform for critical care medicine, especially for multimodal waveform data. Uniquely, it allows algorithms to seamlessly integrate into the live data stream to produce clinical decision support and predictions in clinical practice.

Clinical Evaluation of Antiphospholipase A2 Receptor IgG4 level and Its IgG4-to-IgG Ratio Based on Quantitative Immunoassays in Idiopathic Membranous Nephropathy.

Background Phospholipase A2 receptor (PLA2R), located at the membrane of glomerular podocyte, is the major autoantigen of idiopathic membranous nephropathy (IMN), and its antibodies with a predominant IgG4 subclass lead to pathological lesions. Further studies could be performed to validate the clinical values of PLA2R-IgG, PLA2R-IgG4, and PLA2R-IgG4-to-IgG ratios, as ultrasensitive and quantitative immunoassays for PLA2R antibodies have been well established in our previous work. Methods A cohort of 58 IMN patients, 30 of whom were followed from 3 to 42 months, was assessed for serum PLA2R-IgG and -IgG4 levels, and the ratio of PLA2R-IgG4/-IgG combined with relative clinicopathological indicators. Results Serum PLA2R-IgG4 level was significantly correlated with glomerular PLA2R staining. In addition, it was strongly correlated with PLA2R-IgG and its ratio. PLA2R-IgG and -IgG4 levels were both correlated with high-density lipoprotein and erythrocyte sedimentation rates. The ratio at the first diagnosis can predict the remission, and its efficacy overmatched PLA2R-IgG4. In the survival curves, negative results for the ratio or PLA2R-IgG4 at the first diagnosis demonstrated higher remission rates. Conclusion Serum PLA2R-IgG4 concentration may replace renal PLA2R immunohistochemistry in IMN diagnosis. We propose that the PLA2R-IgG4-to-IgG ratio and PLA2R-IgG4 could be novel indicators for remission prediction in clinical practice.

Implementing Machine Learning Models for Suicide Risk Prediction in Clinical Practice: Focus Group Study With Hospital Providers.

BackgroundInterest in developing machine learning models that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. However, whether and how such models might be implemented and useful in clinical practice remain unknown. To ultimately make automated suicide risk–prediction models useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders, including the frontline providers who will be using such tools, at each stage of the implementation process.ObjectiveThe aim of this focus group study is to inform ongoing and future efforts to deploy suicide risk–prediction models in clinical practice. The specific goals are to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk–prediction models in practice; and identify barriers, facilitators, recommendations, and factors to consider.MethodsWe conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by 2 independent study staff members. All coded text was reviewed and discrepancies were resolved in consensus meetings with doctoral-level staff.ResultsAlthough most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk–prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the health care system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider training. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings.ConclusionsProviders were dissatisfied with current suicide risk assessment methods and were open to the use of a machine learning–based risk-prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of these new approaches in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.

Derivation and Validation of a Clinical Score for Predicting Postoperative Atrial Fibrillation in Noncardiac Elective Surgery (the HART Score)

Postoperative atrial fibrillation (POAF) represents the most frequent cardiac arrhythmia in the surgical setting. It affects almost 3% of all patients over 45 years old who underwent noncardiovascular surgery and is associated with a higher risk of stroke, heart failure, myocardial infarction, and cardiac arrest. The study aimed to assess independent predictors of POAF and derive and validate a score for risk prediction in clinical practice. This was a retrospective cohort study including all consecutive candidates to all types of noncardiac elective surgery attending a cardiological preoperative assessment from 2016 to 2019. Exclusion criteria were a previous diagnosis of AF and the cancelation of the planned surgery. A total of 2,048 patients were enrolled (1350 men, aged 72 ± 12 years). A total of 44 patients experienced POAF (2.1%) - median 3 days (first to third quartile 2 to 4 days). Age (odds ratio [OR] 1.03 for each year, 95% confidence interval [CI] 1.01 to 1.07), hypertension (OR 3.43, 95% CI 1.22 to 9.63), thyroid dysfunction (OR 2.47, 95% CI 1.22 to 5.01), and intermediate or high-risk surgery (OR 18.28, 95% CI 2.51 to 33.09) resulted as independent predictors of POAF (all p <0.05). The Hypertension, Age, surgery Risk, and Thyroid dysfunction score (OR 2.59 for each point, 95% CI 1.79 to 3.75, p <0.001) was then created based on those 4 items. A cut-off score ≥6 had a 70% sensitivity and a 72% specificity in detecting POAF (area under the curve 0.76). Bootstrapping for internal validation confirmed the overall results (area under the curve 0.72). In conclusion, POAF complicates around 2% of all noncardiac surgery. A 4-item risk score, such as the Hypertension, Age, surgery Risk, and Thyroid dysfunction score, could be effective in implementing POAF screening and improving management.

Assessing the utility of deep neural networks in predicting postoperative surgical complications: a retrospective study

Early detection of postoperative complications, including organ failure, is pivotal in the initiation of targeted treatment strategies aimed at attenuating organ damage. In an era of increasing health-care costs and limited financial resources, identifying surgical patients at a high risk of postoperative complications and providing personalised precision medicine-based treatment strategies provides an obvious pathway for reducing patient morbidity and mortality. We aimed to leverage deep learning to create, through training on structured electronic health-care data, a multilabel deep neural network to predict surgical postoperative complications that would outperform available models in surgical risk prediction. In this retrospective study, we used data on 58 input features, including demographics, laboratory values, and 30-day postoperative complications, from the American College of Surgeons (ACS) National Surgical Quality Improvement Program database, which collects data from 722 hospitals from around 15 countries. We queried the entire adult (≥18 years) database for patients who had surgery between Jan 1, 2012, and Dec 31, 2018. We then identified all patients who were treated at a large midwestern US academic medical centre, excluded them from the base dataset, and reserved this independent group for final model testing. We then randomly created a training set and a validation set from the remaining cases. We developed three deep neural network models with increasing numbers of input variables and so increasing levels of complexity. Output variables comprised mortality and 18 different postoperative complications. Overall morbidity was defined as any of 16 postoperative complications. Model performance was evaluated on the test set using the area under the receiver operating characteristic curve (AUC) and compared with previous metrics from the ACS-Surgical Risk Calculator (ACS-SRC). We evaluated resistance to changes in the underlying patient population on a subset of the test set, comprising only patients who had emergency surgery. Results were also compared with the Predictive OpTimal Trees in Emergency Surgery Risk (POTTER) calculator. 5 881 881 surgical patients, with 2941 unique Current Procedural Terminology codes, were included in this study, with 4 694 488 in the training set, 1 173 622 in the validation set, and 13 771 in the test set. The mean AUCs for the validation set were 0·864 (SD 0·053) for model 1, 0·871 (0·055) for model 2, and 0·882 (0·053) for model 3. The mean AUCs for the test set were 0·859 (SD 0·063) for model 1, 0·863 (0·064) for model 2, and 0·874 (0·061) for model 3. The mean AUCs of each model outperformed previously published performance metrics from the ACS-SRC, with a direct correlation between increasing model complexity and performance. Additionally, when tested on a subgroup of patients who had emergency surgery, our models outperformed previously published POTTER metrics. We have developed unified prediction models, based on deep neural networks, for predicting surgical postoperative complications. The models were generally superior to previously published surgical risk prediction tools and appeared robust to changes in the underlying patient population. Deep learning could offer superior approaches to surgical risk prediction in clinical practice. The Novo Nordisk Foundation.

Predicting falls in community-dwelling older adults: a systematic review of prognostic models

ObjectiveTo systematically review and critically appraise prognostic models for falls in community-dwelling older adults.Eligibility criteriaProspective cohort studies with any follow-up period. Studies had to develop or validate multifactorial prognostic models...

Artificial intelligence: Deep learning in oncological radiomics and challenges of interpretability and data harmonization.

Over the last decade there has been an extensive evolution in the Artificial Intelligence (AI) field. Modern radiation oncology is based on the exploitation of advanced computational methods aiming to personalization and high diagnostic and therapeutic precision. The quantity of the available imaging data and the increased developments of Machine Learning (ML), particularly Deep Learning (DL), triggered the research on uncovering "hidden" biomarkers and quantitative features from anatomical and functional medical images. Deep Neural Networks (DNN) have achieved outstanding performance and broad implementation in image processing tasks. Lately, DNNs have been considered for radiomics and their potentials for explainable AI (XAI) may help classification and prediction in clinical practice. However, most of them are using limited datasets and lack generalized applicability. In this study we review the basics of radiomics feature extraction, DNNs in image analysis, and major interpretability methods that help enable explainable AI. Furthermore, we discuss the crucial requirement of multicenter recruitment of large datasets, increasing the biomarkers variability, so as to establish the potential clinical value of radiomics and the development of robust explainable AI models.

The efficiency of prostate-specific antigen density measurement using three different methods on the prediction of biochemical recurrence

Background The aim of this study was to evaluate the efficiency of prostate-specific antigen (PSA) density (PSAD) calculated through prostate volume (PV) obtained via transrectal ultrasound (TRUS) and magnetic resonance imaging (MRI) and actual prostate weight (PW) methods obtained via pathological evaluation on the prediction of biochemical recurrence (BCR) in the follow-ups of patients who had undergone radical prostatectomy (RP). Methods A total of 335 clinically localized prostate cancer (PCa) patients who had received open RP between January 2015 and December 2018 were enrolled in the study. Pre and postoperative demographic data, clinical and pathological findings and BCR conditions were recorded. The PSAD was calculated using information obtained through preoperative TRUS examinations, MRI, and collected pathological specimens after RP by dividing the maximum preoperative PSA value and PV/PW. Results In a mean follow-up duration of 20.2 ± 8.5 months, recurrence was observed in 52 patients (24.4%) and progression was observed in 8 (3.8%) patients. The TRUS-PSAD, MRI-PSAD, and PW-PSAD values were statistically significantly higher in BCR patients compared to non-BCR patients. The International Society of Urologic Pathologists (ISUP) grade 5 and pT3b as a pathological stage were detected as independent variables in the prediction of BCR formation. Actual PW had a high prediction value compared to other PSAD measurements at <40 g prostate weights, but it had a low prediction value in prostates with an actual PW >60 g. Conclusions In this study, it was stated that PSAD acquired through different imaging methods does not affect the usability of PSAD in BCR prediction in clinical practice. The ISUP grade 5 and pT3b stage PCa were detected as independent markers in BCR prediction after RP.

3D Analysis of the Proximal Femur Compared to 2D Analysis for Hip Fracture Risk Prediction in a Clinical Population

Due to the adverse impacts of hip fractures on patients' lives, it is crucial to enhance the identification of people at high risk through accessible clinical techniques. Reconstructing the 3D geometry and BMD distribution of the proximal femur could be beneficial in enhancing hip fracture risk predictions; however, it is associated with a high computational burden. It is also not clear whether it provides a better performance than 2D model analysis. Therefore, the purpose of this study was to compare the 2D and 3D model reconstruction's ability to predict hip fracture risk in a clinical population of patients. The DXA scans and CT scans of 16 cadaveric femurs were used to create training sets for the 2D and 3D model reconstruction based on statistical shape and appearance modeling. Subsequently, these methods were used to predict the risk of sustaining a hip fracture in a clinical population of 150 subjects (50 fractured, and 100 non-fractured) that were monitored for five years in the Canadian Multicentre Osteoporosis Study. 3D model reconstruction was able to improve the identification of patients who sustained a hip fracture more accurately than the standard clinical practice (by 40%). Also, the predictions from the 2D statistical model didn't differ significantly from the 3D ones (p > 0.76). These results indicated that to enhance hip fracture risk prediction in clinical practice implementing 2D statistical modeling has comparable performance with lower associated computational load.

Admission blood cell counts are predictive of stroke-associated infection in acute ischemic stroke patients treated with endovascular therapy.

Stroke-associated infection (SAI) is a major medical complication in acute ischemic stroke patients (AIS) treated with endovascular therapy (EVT). Three hundred thirty-three consecutive patients with AIS caused by a large vessel occlusion in the anterior circulation who received EVT (142 (42.6%) of them were given IV tPA as bridging therapy) and 337 AIS patients who received IV tPA only (non-EVT) were enrolled in the study and evaluated to determine the association of inflammatory factors on admission with SAI. Among the 333 AIS patients undergoing EVT, SAI occurred in 219 (65.8%) patients. Patients with SAI had higher baseline National Institutes of Health Stroke Scale (NIHSS) total scores, white blood cell (WBC) and neutrophil counts, neutrophil-to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR) than those without SAI (P < 0.05). The multivariable logistic regression analyses showed that older age in addition to higher diastolic blood pressure (DBP), NIHSS score, fasting blood glucose, WBC and neutrophil counts, NLR, and PLR were significantly associated with SAI (P < 0.05). However, these associations were not revealed in 337 non-EVT AIS patients. Furthermore, based on the inflammatory markers, we developed a nomogram that provided the opportunity for more accurate predictions (compared with conventional factors) and appeared a better prognostic tool for SAI according to the decision curve analysis. In summary, if proven externally valid, our nomogram that included WBC count, NLR, and PLR may be a useful tool for SAI prediction in clinical practice.

Weighted Epworth sleepiness scale predicted the apnea-hypopnea index better

BackgroundThe relationship between the Epworth sleepiness scale (ESS) and the apnea-hypopnea index (AHI) is uncertain and even poor. The major problem associated with the ESS might be a lack of consideration of weight in prediction in clinical practice. Would awarding different item-scores to the four scales of ESS items to develop a weighted ESS scoring system improve the accuracy of the AHI prediction? It is warranted to explore the intriguing hypotheses.MethodsSeven hundred fifty-six adult patients with suspicion of obstructive sleep apnoea syndrome (OSAS) were prospectively recruited to a derivation cohort. This was tested against a prospective validation cohort of 810 adult patients with suspected OSAS. Each ESS item’s increased odds ratio for the corresponding AHI was calculated using univariate logistic regression. The receiver operating characteristic curves were created and the areas under the curves (AUCs) were calculated to illustrate and compare the accuracy of the indices.ResultsThe higher the ESS item-score, the closer the relationship with the corresponding AHI. The odds ratios decreased as a result of the increased AHI. The ESS items were of unequal weight in predicting the corresponding AHI and a weighted ESS was developed. The coincidence rates with the corresponding AHI, body mass indices, and neck circumferences rose as the scores increased, whereas nocturnal nadir oxygen saturations decreased, and the weighted ESS was more strongly associated with these indices, compared with the ESS. The capability in predicting the patients without OSAS or with severe OSAS was strong, especially the latter, and the weighted ESS orchestrated manifest improvement in screening the patients with simple snoring. The patterns of sensitivities, specificities, and Youden’s indices of the four ranks of weighted ESS for predicting the corresponding AHI were better than those of the ESS, and the AUCs of weighted ESS were greater than the corresponding areas of ESS in the two cohorts.ConclusionsThe weighted ESS orchestrated significant improvement in predicting the AHI, indicating that the capability in predicting the patients without OSAS or with severe OSAS was strong, which might have implications for clinical triage decisions to prioritize patients for polysomnography.

Nonlinear voxel-based finite element model for strength assessment of healthy and metastatic proximal femurs

Nonlinear finite element (FE) models can accurately quantify bone strength in healthy and metastatic femurs. However, their use in clinical practice is limited since state-of-the-art implementations using tetrahedral meshes involve a lot of manual work for which specific modelling software and engineering knowledge are required. Voxel-based meshes could enable the transition since they are robust and can be highly automated. Therefore, the aim of this work was to bridge the modelling gap between the tetrahedral and voxel-based approach. Specifically, we validated a nonlinear voxel-based FE method relative to experimental data from 20 femurs with and without artificial metastases that had been mechanically loaded until failure. CT scans of the femurs were segmented and automatically converted into a voxel-based mesh with hexahedral elements. Nonlinear material properties were implemented in an open-source linear voxel-based FE solver by adding an additional loop to the routine such that the material properties could be adapted after each increment. Bone strength, quantified as the maximum force in the force-displacement curve, was evaluated. The results were compared to a previously established nonlinear tetrahedral FE approach as well as to the experimentally measured bone strength. The voxel-based FE model predicted the experimental bone strength very well both for healthy (R2 = 0.90, RMSE = 0.88 kN) and metastatic femurs (R2 = 0.93, RMSE = 0.64 kN). The model precision and accuracy were very similar to the ones obtained with the tetrahedral model (R2 = 0.90/0.93, RMSE = 0.90/0.64 kN for intact/metastatic respectively). The more intuitive voxel-based meshes thus quantified macroscale femoral strength equally well as state-of-the-art tetrahedral models. The robustness, high level of automation and time-efficiency (< 30 min) of the implemented workflow offer great potential for developing FE models to improve fracture risk prediction in clinical practice.

Lipidomic profiling of amniotic fluid and its application in fetal lung maturity prediction.

BackgroundThe pulmonary surfactant especially lipids in amniotic fluid can reflect the development stage of fetal lung maturity (FLM). However, the conventional lecithin/sphingomyelin (L/S) ratio method by thin layer chromatography (TLC) is insufficient and inconvenient for FLM prediction in clinical practice.MethodsThe amniotic fluid samples were collected from the pregnant women in labor or undergoing amniocentesis and analyzed for its lipid contents with the liquid chromatography coupled with high‐resolution mass spectrometry (LC‐HRMS) method and the lamellar body count (LBC) method. To reveal the lipidomic profiling of different FLM stages, three groups of amniotic fluid samples including 8 from premature group (gestational week (GW) < 37), 10 from mature group (GW < 37), and 10 from mature group (GW > 38) were compared with the control group (n = 6) of 18 GWs separately.ResultsIn the FLM prediction study, the sensitivity of the LC‐HRMS method and LBC method was 91% and 73%, respectively; the specificity was 100% and 95%, respectively. The most significant metabolic pathway was linoleic acid metabolism between the premature group and the control group. Both glycerophospholipid metabolism and glycosylphosphatidylinositol‐anchor biosynthesis were enriched in the mature groups. In search of potential FLM prediction markers in amniotic fluid, 8 phosphatidylcholines, 1 sphingomyelin, and 1 phosphatidylethanolamine were significantly increased in the mature groups compared with the premature group.ConclusionAn efficient LC‐HRMS method for L/S ratio in predicting FLM was established. The linoleic acid metabolism may play an important role in the fetal lung development.

Design and Assessment of a Robust and Generalizable ANN-Based Classifier for the Prediction of Premature Birth by means of Multichannel Electrohysterographic Records

Preterm labor is one of the major causes of neonatal deaths and also the cause of significant health and development impairments in those who survive. However, there are still no reliable and accurate tools for preterm labor prediction in clinical settings. Electrohysterography (EHG) has been proven to provide relevant information on the labor time horizon. Many studies focused on predicting preterm labor by using temporal, spectral, and nonlinear parameters extracted from single EHG recordings. However, multichannel analysis, which includes information from the whole uterus and about coupling between the recording areas, may provide better results. The cross validation method is often used to design classifiers and evaluate their performance. However, when the validation dataset is used to tune the classifier hyperparameters, the performance metrics of this dataset may not properly assess its generalization capacity. In this work, we developed and compared different classifiers, based on artificial neural networks, for predicting preterm labor using EHG features from single and multichannel recordings. A set of temporal, spectral, nonlinear, and synchronization parameters computed from EHG recordings was used as the input features. All the classifiers were evaluated on independent test datasets, which were never “seen” by the models, to determine their generalization capacity. Classifiers’ performance was also evaluated when obstetrical data were included. The experimental results show that the classifier performance metrics were significantly lower in the test dataset (AUC range 76-91%) than in the train and validation sets (AUC range 90-99%). The multichannel classifiers outperformed the single-channel classifiers, especially when information was combined into mean efficiency indexes and included coupling information between channels. Including obstetrical data slightly improved the classifier metrics and reached an AUC of 91.1±2.5% for the test dataset. These results show promise for the transfer of the EHG technique to preterm labor prediction in clinical practice.

Prognostic Model for Survival and Recurrence in Patients with Early-Stage Cervical Cancer: A Korean Gynecologic Oncology Group Study (KGOG 1028)

PurposeWe aimed to develop and validate individual prognostic models in a large cohort of cervical cancer patients that were primarily treated with radical hysterectomy.Materials and MethodsWe analyzed 1,441 patients with early-stage cervical cancer treated between 2000 and 2008 from the Korean Gynecologic Oncology Group multi-institutional cohort: a train cohort (n=788) and a test cohort (n=653). Models predicting the risk for overall survival (OS), disease-free survival (DFS), lymphatic recurrence and hematogenous recurrence were developed using Cox analysis and stepwise backward selection and best-model options. The prognostic performance of each model was assessed in an independent patient cohort. Model-classified risk groups were compared to groups based on traditional risk factors.ResultsIndependent risk factors for OS, DFS, lymphatic recurrence, and hematogenous recurrence were identified for prediction model development. Different combinations of risk factors were shown for each outcome with best predictive value. In train cohort, area under the curve (AUC) at 2 and 5 years were 0.842/0.836 for recurrence, and 0.939/0.882 for OS. When applied to a test cohort, the model also showed accurate prediction result (AUC at 2 and 5 years were 0.799/0.723 for recurrence, and 0.844/0.806 for OS, respectively). The Kaplan-Meier plot by proposed model-classified risk groups showed more distinctive survival differences between each risk group.ConclusionWe developed prognostic models for OS, DFS, lymphatic and hematogenous recurrence in patients with early-stage cervical cancer. Combining weighted clinicopathologic factors, the proposed model can give more individualized predictions in clinical practice.

Risk Prediction in Clinical Practice: A Practical Guide for Cardiothoracic Surgeons

Risk Prediction in Clinical Practice: A Practical Guide for Cardiothoracic Surgeons

Discrimination of osteoporosis-related vertebral fractures by DXA-derived 3D measurements: a retrospective case-control study.

The aim of the present study was to evaluate the association of DXA-derived 3D measurements at the lumbar spine with osteoporosis-related vertebral fractures. We retrospectively analyzed a database of 74 postmenopausal women: 37 subjects with incident vertebral fractures and 37 age-matched controls without any type of fracture. DXA scans at the lumbar spine were acquired at baseline (i.e., before the fracture event for subjects in the fracture group), and areal bone mineral density (aBMD) was measured. DXA-derived 3D measurements, such as volumetric BMD (vBMD), were assessed using a DXA-based 3D modeling software (3D-SHAPER). vBMD was computed at the trabecular, cortical, and integral bone. Cortical thickness and cortical surface BMD were also measured. Differences in DXA-derived measurements between fracture and control groups were evaluated using unpaired t test. Odds ratio (OR) and area under the receiver operating curve (AUC) were also computed. Subgroup analyses according to fractured vertebra were performed. aBMD of fracture group was 9.3% lower compared with control group (p < 0.01); a higher difference was found for trabecular vBMD in the vertebral body (- 16.1%, p < 0.001). Trabecular vBMD was the measurement that best discriminates between fracture and control groups, with an AUC of 0.733, against 0.682 for aBMD. Overall, similar findings were observed within the subgroup analyses. The L1 vertebral fractures subgroup had the highest AUC at trabecular vBMD (0.827), against aBMD (0.758). This study showed the ability of cortical and trabecular measurements from DXA-derived 3D models to discriminate between fracture and control groups. Large cohorts need to be analyzed to determine if these measurements could improve fracture risk prediction in clinical practice.

Scored minor criteria for severe community-acquired pneumonia predicted better

BackgroundInfectious Disease Society of America/American Thoracic Society (IDSA/ATS) minor criteria for severe community-acquired pneumonia (CAP) are of unequal weight in predicting mortality, but the major problem associated with IDSA/ATS minor criteria might be a lack of consideration of weight in prediction in clinical practice. Would awarding different points to the presences of the minor criteria improve the accuracy of the scoring system? It is warranted to explore this intriguing hypothesis.MethodsA total of 1230 CAP patients were recruited to a retrospective cohort study. This was tested against a prospective two-center cohort of 1749 adults with CAP. 2 points were assigned for the presence of PaO2/FiO2 ≤ 250 mmHg, confusion, or uremia on admission and 1 point for each of the others.ResultsThe mortality rates, and sequential organ failure assessment (SOFA) and pneumonia severity index (PSI) scores increased significantly with the numbers of IDSA/ATS minor criteria present and minor criteria scores. The correlations of the minor criteria scores with the mortality rates were higher than those of the numbers of IDSA/ATS minor criteria present. As were the correlations of the minor criteria scores with SOFA and PSI scores, compared with the numbers of IDSA/ATS minor criteria present. The pattern of sensitivity, specificity, positive predictive value, and Youden’s index of scored minor criteria of ≥2 scores or the presence of 2 or more IDSA/ATS minor criteria for prediction of mortality was the best in the retrospective cohort, and the former was better than the latter. The validation cohort confirmed a similar pattern. The area under the receiver operating characteristic curve of scored minor criteria was higher than that of IDSA/ATS minor criteria in the retrospective cohort, implying higher accuracy of scored version for predicting mortality. The validation cohort confirmed a similar paradigm.ConclusionsScored minor criteria orchestrated improvements in predicting mortality and severity in patients with CAP, and scored minor criteria of ≥2 scores or the presence of 2 or more IDSA/ATS minor criteria might be more valuable cut-off value for severe CAP, which might have implications for more accurate clinical triage decisions.