Prognostic Algorithms Research Articles

Survival in breast cancer patients with spine metastases: Prognostic assessment involving molecular markers

BackgroundTo clarify and update the prognostic assessment for heterogeneous population of patients with breast cancer and spine metastases (SpM), using molecular markers. MethodsThe patient data used in this study was obtained from a French national multi-center database of patients treated for breast cancer with SpM between 2014 and 2017. 556 SpM cases were diagnosed. ResultsMedian overall survival (OS) time for all patients following the SpM event was 43.9 months. First, we confirmed 3 previously known significant prognostic factors for survival of patients with SpM: young age [HR: 2.019, 95% CI 1.343–3.037; p = 0.001], good WHO status [ Status 0 HR: 2.823, 95% CI 1.231–3.345; p < 0.0001] or [ Status 1 HR: 1.956, 95% CI 0.768–2.874; p = 0.001] and no-ambulatory neurological status: Frankel A-C [HR: 0.438, 95% CI 0.248–0.772; p = 0.004]. Secondly, we determined the effect of gene mutations on survival in patients with SpM, and we identified that HER2+ cancer subtype [HR: 1.567, 95% CI 0.946–2.557; p = 0.008] was an independent predictor of longer survival, whereas basal cancer subtype [HR: 0.496, 95% CI 0.353–0.699; p < 0.0001] was associated with a poorer prognosis. Other factors including the number of SpM, surgery, extraspinal metastases, synchrone metastases, metastasis-free survival, and SpM recurrence were not identified as prognostically relevant to survival. ConclusionSurvival and our ability to estimate it in breast cancer patients with SpM has improved significantly. Therefore, SpM prognostic scoring algorithms should be updated and incorporate genotypic data on subtypes to make treatment more adaptive.

Создание алгоритма прогнозирования сосудистой ретинальной патологии у женщин после перенесенной преэклампсии и оценка его эффективности

Purpose. On the basis of previously identified potential predictors, to develop an algorithm for a prediction of vascular retinal pathology in women who have previously suffered from pregnancy complicated by preeclampsia. Material and methods. In the study there was conducted A dynamic observation of 83 women (133 eyes), who suffered from preeclampsia, ending in childbirth. The age of patients at the time of pregnancy ranged from 18 to 45 years, mean 31.9±6.1 years. In addition to the standard eye examination, all women underwent a study of chorioretinal hemodynamics and morphometric parameters of the macular area. Results. Based on previously revealed potential predictors, three variants of the algorithm for prediction of vascular retinal pathology in women after preeclampsia were developed. Algorithms can be considered reasonable and effective. They can be recommended to clinicians for the selection of women suffered from preeclampsia in the group with a risk of vascular retinal pathology development in long-term periods after childbirth. Conclusion. Three variants of the prognostic algorithm of risk for the formation of vascular retinal pathology in women after suffered preeclampsia have been developed, for periods up to 10 years. They differ from each other in various sets of predictors and are oriented, respectively, for the use in the following conditions: specialized ophthalmologic clinic, ophthalmology office of the out-patient clinic and obstetriciangynecologists.

Combining gene variants with clinical characteristics improves outcome prediction in Chinese patients with myelodysplastic syndromes

Genetic variants have been identified in the majority of myelodysplastic syndromes (MDS) patients and have considerably influenced the diagnosis, classification, risk stratification and treatment of MDS. To explore the prognostic significance of genomic variants and build a new prognostic scoring model, we performed next-generation sequencing of 51 known genes in 499 Chinese patients with MDS. Ultimately, the TP53, GATA2, DNMT3A, age and the revised International Prognostic Scoring System (IPSS-R) risk stratification were included in a new Cox model and divided into three prognostic categories, and had a better prediction of overall survival. The C-index of the new prognostic scoring model (0.772) was clearly better than IPSS-R risk stratification (0.717), which was validated in 163 cases. Moreover, the new model was also suitable for the prediction of OS for patients undergoing allogeneic hematopoietic stem cell transplantation. The inclusion of genomic variants and age into the IPSS-R could improve prognostic algorithms for MDS patients.

Comparison of seven prognostic tools to identify low-risk pulmonary embolism in patients aged

In young patients with acute pulmonary embolism (PE), the predictive value of currently available prognostic tools has not been evaluated. Our objective was to compare prognostic value of 7 available tools (GPS, PESI, sPESI, Prognostic Algorithm, PREP, shock index and RIETE) in patients aged <50 years. We used the RIETE database, including PE patients from 2001 to 2017. The major outcome was 30-day all-cause mortality. Of 34,651 patients with acute PE, 5,822 (17%) were aged <50 years. Of these, 83 (1.4%) died during the first 30 days. Number of patients deemed low risk with tools was: PREP (95.9%), GPS (89.6%), PESI (87.2%), Shock index (70.9%), sPESI (59.4%), Prognostic algorithm (58%) and RIETE score (48.6%). The tools with a highest sensitivity were: Prognostic Algorithm (91.6%; 95% CI: 85.6–97.5), RIETE score (90.4%; 95%CI: 84.0–96.7) and sPESI (88%; 95% CI: 81–95). The RIETE, Prognostic Algorithm and sPESI scores obtained the highest overall sensitivity estimates for also predicting 7- and 90-day all-cause mortality, 30-day PE-related mortality, 30-day major bleeding and 30-day VTE recurrences. The proportion of low-risk patients who died within the first 30 days was lowest using the Prognostic Algorithm (0.2%), RIETE (0.3%) or sPESI (0.3%) scores. In PE patients less 50 years, 30-day mortality was low. Although sPESI, RIETE and Prognostic Algorithm scores were the most sensitive tools to identify patients at low risk to die, other tools should be evaluated in this population to obtain more efficient results.

Evaluating Whether Integration of Prognostic and Predictive Algorithms Into Routine Clinical Practice Effect Whether Oncologists Order Multigene Assays in Patients With Early Stage Breast Cancer

Evaluating Whether Integration of Prognostic and Predictive Algorithms Into Routine Clinical Practice Effect Whether Oncologists Order Multigene Assays in Patients With Early Stage Breast Cancer

A method for the reduction of the computational cost associated with the implementation of particle-filter-based failure prognostic algorithms

Abstract Failure prognostic algorithms require to reduce the computational burden associated with their implementation to ensure real-time performance in embedded systems. In this regard, this paper presents a method that allows to significantly reduce this computational cost in the case of particle-filter-based prognostic algorithms, which is based on a time-variant prognostic update rate. In this proposed scheme, the performance of the prognostic algorithm within short-term prediction horizons is continuously compared with respect to the outcome of Bayesian state estimators. Only if the discrepancy between prior and posterior knowledge is greater than a given threshold, it is suggested to execute the prognostic algorithm once again and update Time-of-Failure estimates. In addition, a novel metric to evaluate the performance of any prognostic algorithm in real-time is hereby presented. The proposed actualization scheme is implemented, tested, and validated in two case studies related to the problem of State-of-Charge (SOC) prognostics. The obtained results show that the proposed strategy allows to significantly reduce the computational cost while keeping the standards in terms of algorithm efficacy.

Relationship of Serum Antileishmanial Antibody With Development of Visceral Leishmaniasis, Post-kala-azar Dermal Leishmaniasis and Visceral Leishmaniasis Relapse.

IntroductionTo sustain the achievement of kala-azar elimination program (KEP), early detection and treatment of the visceral leishmaniasis (VL) cases and associated modalities such as treatment failure (TF), relapse VL (RVL), and Post-kala-azar dermal leishmaniasis (PKDL) is the cornerstone. A predictive biomarker for VL development and related complications could also play a crucial role in curtailing disease incidence and transmission. Investigations to find a biomarker with prospective capabilities are, however, scarce. Using samples and known clinical outcomes generated within two previous longitudinal cohort studies, we aimed to determine if fluctuations in serum anti-rK39 antibody levels could provide such predictive value.Materials and MethodsSerum samples collected at four different time points (Baseline, 12, 18, and 24 months) from 16 patients who had developed VL within the monitoring period and 15 of their asymptomatic healthy controls counterparts were investigated. To investigate potential prediction of VL related complications, serum samples of 32 PKDL, 10 RVL, 07 TF, and 38 cured VL from a single dose AmBisome trial were analyzed. Of this second panel, all patients were monitored for 5 years and sera were collected at four time points (Baseline then 1, 6, and 12 months after treatment). The level of anti-rK39 antibodies in archived samples was measured by a semi-quantitative ELISA.ResultsThe mean antibody level was significantly higher in VL patients compared to their asymptomatic healthy counterparts at each time point. Likewise, we observed a trend toward elevations in antibody levels for PKDL, RVL, TF relative to the reducing levels observed in cured VL. Receiver operating characteristic (ROC) analysis found a promising predictive power of rK39 antibody levels to reveal progression from asymptomatic Leishmania donovani infection stage to VL, defined as 87.5% sensitive and 95% specific. Following treatment, rk39 antibody notably showed 100% sensitivity and 95% specificity in predicting TF.ConclusionOur data indicate that the relative quantity of serum anti-rK39 antibody has promise within either a predictive or prognostic algorithm for VL and VL-related modalities. These could enable VL control programs to implement more effective measures to eliminate the disease. Further research is, however, imperative to standardize the rK39 antibody ELISA between sites prior to broader use.

The diagnostic value of using nomograms in predicting the occurrence of colorectal anastomosis leakage (a literature review)

The anastomosis leakage is one of the most terrible complications of rectal cancer surgery. The anastomosis leakage rate after low anterior resection may exceed 20 %. The anastomosis leakage increases postoperative mortality, length of hospitalization, requires the formation of a temporary or permanent colostomy, which impairs the quality of life of patients. In recent years, some authors have suggested using a nomogram, a prognostic algorithm in the form of a graphic image showing the probability of a certain outcome individually for each patient. This review presents literature data on the use of nomograms in predicting the occurrence of colorectal anastomosis leakage after rectal cancer surgery.

Synovial tissue signatures enhance clinical classification and prognostic/treatment response algorithms in early inflammatory arthritis and predict requirement for subsequent biological therapy: results from the pathobiology of early arthritis cohort (PEAC)

ObjectiveTo establish whether synovial pathobiology improves current clinical classification and prognostic algorithms in early inflammatory arthritis and identify predictors of subsequent biological therapy requirement.Methods200 treatment-naïve patients with early arthritis were...

A prognostic algorithm to prescribe improvement measures on throughput bottlenecks

Throughput bottleneck analysis is important in prioritising production and maintenance measures in a production system. Due to system dynamics, bottlenecks shift between different production resources and across production runs. Therefore, it is important to predict where the bottlenecks will shift to and understand the root causes of predicted bottlenecks. Previous research efforts on bottlenecks are limited to only predicting the shifting location of throughput bottlenecks; they do not give any insights into root causes. Therefore, the aim of this paper is to propose a data-driven prognostic algorithm (using the active-period bottleneck analysis theory) to forecast the durations of individual active states of bottleneck machines from machine event-log data from the manufacturing execution system (MES). Forecasting the duration of active states helps explain the root causes of bottlenecks and enables the prescription of specific measures for them. It thus forms a machine-states-based prescriptive approach to bottleneck management. Data from real-world production systems is used to demonstrate the effectiveness of the proposed algorithm. The practical implications of these results are that shop-floor production and maintenance teams can be forewarned, before a production run, about bottleneck locations, root causes (in terms of machine states) and any prescribed measures, thus forming a prescriptive approach. This approach will enhance the understanding of bottleneck behaviour in production systems and allow data-driven decision making to manage bottlenecks proactively.

978P - Algorithms derived from quantitative pathology can be a gatekeeper in patient selection for clinical trials in localised clear cell renal cell carcinoma (ccRCC)

BackgroundIdentifying low risk of disease recurrence in localised ccRCC is key for gatekeeping in the adjuvant therapy enrolment. Uncertainty increases the number of patients required for accrual to achieve statistical power. Current scoring systems are good at identifying very low and very high risk cohorts, but have not been proved to be as effective at accurately predicting disease recurrence in intermediate groups. These patients are perhaps those likely to benefit from intervention in addition to surgery, but many may be treated unnecessarily. Using digital pathology, image analysis and machine learning we sought to stratify for risk in this intermediate category. MethodsDefiniens Tissue studio and Developer XD were utilised for object thresholding to measure tumour cell nuclear morphological features on H&E digitised images from 120 ccRCC training set from UK and 217 ccRCC validation set from Singapore. Multiplexed immunofluorescence (mIF) was performed on 120 cases to co-detect neo-vasculature and pan-T cells. An algorithm was derived to measure spatial relationships between blood vessels and T cells. A statistical model was developed by generalised linear model with spatially adaptive local smoothing algorithm, having specificity prefixed (0.8-1) plus cross validation. ResultsReplacing manual nuclear grade with AI aided tumour cell nuclear morphological features improved the specificity of Leibovich score (LS) from 0.76 to 0.86 and from 0.84 to 0.94 in training and validation sets, respectively. Moreover, tumour microenvironment (TM) parameters significantly improved the specificity up to 0.93 in the training set. The negative predictive values of both LS 5 and 6 were zero, but by applying the algorithm the specificity for LS 5 and 6 cases became 0.93 and 0.40 respectively. ConclusionsBy applying image analysis it is possible to identify lower risk for recurrence patients in a conventionally identified intermediate risk group based on routine ccRCC H&E images, and multiplexed TM features. This approach to pathology should help refine selection of patients for clinical trials and form the basis of future AI-enabled prognostic and predictive algorithms in ccRCC. Legal entity responsible for the studyAuthors. FundingRenal Cancer Research Fund and NHS Lothian. DisclosureAll authors have declared no conflicts of interest.

P2767Outcomes and incremental prognostic value of renal function impairment after acute pulmonary embolism

Abstract Background We explored the relation between adverse outcomes after acute pulmonary embolism (PE)and renal dysfunction classified by estimated glomerular filtration rate (eGFR) using the CKD-EPI equation. We assessed the incremental value of adding eGFR CKD-EPI to the ESC score for predicting 30d mortality. Methods Prospective, multicenter study of 1664 acute PE admitted from 01/2011 to 12/2017. Pts were categorized in 4 eGFR groups: Group 1 (eGFR ≥60 ml/min/1.73m2, n=1178), group 2 (45–59; n=257); group 3 (30–44; n=150), group 4 (≤29; n=79). Results All-cause and CV death at 30 days and 6 months were higher in group 3 (p=0.005 and p=0.03) and group 4 (p&lt;0.001 and p&lt;0.001 respectively) vs group 1. Major bleeding at 30d and 6m was higher in group 2 vs group 1 (p=0.003 for both). Renal dysfunction combined with the ESC prognostic algorithm for prediction of 30d death improved discriminatory capacity of the model and enabled reclassification in different risk categories in 27% of pts (Table). Without eGFR CKD-EPI OR (95% CI) With eGFR CKD-EPI OR (95% CI) ESC algorithm 2.59 (1.95–3.43) 2.30 (1.72–3.07) eGFR CKD-EPI – 2.60 (1.62–4.7) Measures of fit Bayes information criterion 607.30 599.32 Akaike information criterion 596.47 583.06 C-statistic 0.71* 0.77* P (Hosmer-Lemeshow) 0.057 0.43 Integrated discrimination improvement – 0.054 (0.052–0.056) Net reclassification improvement – 0.93 (0.90–0.95) Prognostic performance Sensitivity 62.5 (51.2–72.3) 76.2 (61.5–90.2) Specificity 64.2 (49.1–74.4) 69.9 (47.6–83.5) Positive predictive value 10.1 (8.2–11.3) 16.2 (14.2–18.2) Negative predictive value 0.97 (0.96–0.98) 98.1 (97.2–99.2) Positive likelihood ratio 1.96 (1.12–3.41) 2.12 (1.54–3.12) Negative likelihood ratio 0.50 (0.25–1.81) 0.54 (0.20–1.56) Youden index 0.31 (0.28–0.34) 0.39 (0.36–0.41) Difference in C-statistic: *p=0.04. Conclusion Renal function impairment increases the rate of adverse events after acute PE. Combined with the ESC early mortality risk score, eGFR improves risk classification.

16 Model for Predicting Severity in Acute Pancreatitis

INTRODUCTION: Acute pancreatitis (AP) is severe in about 20% of cases with an associated mortality as high as 40%. Accurate prediction of AP severity is necessary to both triage patients and to identify those that might require specific early medical management (EMM). Existing AP severity scoring methods are based on retrospective data and derived from cohorts managed without standardized EMM. The aim of this study was to identify a prognostic model to predict AP severity in a prospective cohort using an evidence-based EMM protocol. METHODS: All patients diagnosed with AP from 1/2015–9/2018 were managed per protocol that included emergency department and admit order sets to standardize EMM, e.g., early aggressive fluid resuscitation and enteral feeding. Exclusion criteria included age less than 18 yrs, transfers from outside hospitals, post-ERCP AP, and trauma related AP. Final severity of AP was categorized as either mild or severe (severe and moderately severe) as per Revised Atlanta Classification. Univariate analysis was performed to identify variables of interest. Classification and regression tree (CART®) analysis (a non-parametric method based on the binary recursive partitioning of data) was used to identify the baseline variables most important in predicting severity, determine cutoffs, and to create a decision tree model. Class probability function was used as the splitting rule for tree building with the limit set to not split the node if sample &lt;30 and that the terminal node not have sample &lt;10. We used 20-fold cross validation method for validating the tree model, the overall performance of which was validated in the test sample using area under the receiver operating characteristic (ROC) curve. RESULTS: 417 total patients were included in the cohort; 356 and 61 patients met criteria for mild and severe pancreatitis, respectively. CART analysis identified five baseline variables in order of greatest to least importance: Creatinine (Cr), procalcitonin, SIRS, age, and HCT (Figure 1). The analysis identified a prognostic algorithm using cutoffs to predict AP severity (Figure 2). ROC curves were generated for the learned (83.7%) and test (73.6%) data sets, respectively (Figure 3). CONCLUSION: We identified baseline variables and an algorithm to predict AP severity from a prospective cohort managed using a standardized, evidence based approach. Our results should be validated prospectively to generate a simple scoring model that can be used to appropriately care for patients with AP.

Prognostics As-A-Service: A Scalable Cloud Architecture for Prognostics

&#x0D; &#x0D; &#x0D; Comprehensive aircraft system health-state awareness is crit- ical for maintaining safe, efficient growth in global oper- ations, enabling higher levels of autonomy, and facilitat- ing new forms of aviation. Maintainers, vehicle operators, air traffic controllers, dispatchers, pilots, autonomous sys- tems, and other decision-makers must have reliable real-time knowledge of the vehicle health, the health of its critical com- posite systems, predictions of how health changes with time, and forecasts of how its capabilities change with health degra- dation to preserve safety and efficiency. Providing this infor- mation in a reliable manner in computationally constrained environments and across a wide range of vehicles and sys- tems continues to be a challenge. This challenge can be par- tially resolved through cloud computing, where the execu- tion of prognostic and diagnostic algorithms is performed on a network of remote servers hosted on the internet. NASA is developing a cloud computing service, Prognostics As-A- Service (PaaS), that explores the feasibility and challenges of cloud-enhanced prognostics. Though such a system has broad applicability, this research effort is focused on aviation applications.&#x0D; &#x0D; &#x0D;

Beyond Survival in Solid Organ Transplantation: A Summary of Expert Presentations from the Sandoz 6th Standalone Transplantation Meeting, 2018.

Beyond Survival in Solid Organ Transplantation: A Summary of Expert Presentations from the Sandoz 6th Standalone Transplantation Meeting, 2018.

Intelligent data-driven prognostic methodologies for the real-time remaining useful life until the end-of-discharge estimation of the Lithium-Polymer batteries of unmanned aerial vehicles with uncertainty quantification

In this paper, the discharge voltage is utilized as a critical indicator towards the probabilistic estimation of the Remaining Useful Life until the End-of-Discharge of the Lithium-Polymer batteries of unmanned aerial vehicles. Several discharge voltage histories obtained during actual flights constitute the in-house developed training dataset. Three data-driven prognostic methodologies are presented based on state-of-the-art as well as innovative mathematical models i.e. Gradient Boosted Trees, Bayesian Neural Networks and Non-Homogeneous Hidden Semi Markov Models. The training and testing process of all models is described in detail. Remaining Useful Life prognostics in unseen data are obtained from all three methodologies. Beyond the mean estimates, the uncertainty associated with the point predictions is quantified and upper/lower confidence bounds are also provided. The Remaining Useful Life prognostics during six random flights starting from fully charged batteries are presented, discussed and the pros and cons of each methodology are highlighted. Several special metrics are utilized to assess the performance of the prognostic algorithms and conclusions are drawn regarding their prognostic capabilities and potential.

Valve Failure Prognostics in Reciprocating Compressors Utilizing Temperature Measurements, PCA-Based Data Fusion, and Probabilistic Algorithms

In this paper, temperature measurements are utilized to develop health indicators based on principal component analysis toward the probabilistic estimation of the remaining useful life (RUL) of reciprocating compressors in service. Temperature degradation histories obtained from 13 actual valve failure cases constitute the training data in a data-driven prognostic approach. Two data-driven prognostic methodologies are presented and proposed based on probabilistic mathematical models, i.e., gradient boosted trees and nonhomogeneous hidden semi-Markov models. The training and testing process of all models is described in detail. RUL prognostics in unseen data are obtained for all models. Beyond the mean estimates of the RUL, the uncertainty associated with the point prediction is quantified and upper/lower confidence bounds are also estimated. Prediction estimates for 12 real-life failure cases are presented and the pros and cons of each model's performance are highlighted. Several metrics are utilized to assess the performance of the prognostic algorithms and conclusions are drawn regarding the prognostic capabilities of each of them.

Piecewise model based intelligent prognostics for state of health prediction of rechargeable batteries with capacity regeneration phenomena

State of health prediction of rechargeable batteries is an important topic in battery health management systems to infer remaining charge-discharge cycles (RCDC) and ensure high reliability of rechargeable batteries. To investigate various prognostic algorithms, battery life-cycle fade experiments were conducted in the NASA Ames Prognostic Center of Excellence. Even though some prognostic algorithms were proposed in the past years to predict RCDC, most of them omitted the influence of battery capacity regeneration phenomena (CRP) on RCDC prediction. Battery CRP cause some sudden capacity increments at unexpected charge-discharge cycles, which may further result in inaccurate RCDC prediction if such CRP are not considered. In this paper, piecewise model based intelligent prognostics for RCDC prediction of rechargeable batteries with CRP is proposed. Results show that the proposed prognostic model can precisely estimate NASA battery capacity fade data, automatically detect CRP and accurately predict RCDC.

Remaining Useful Life Prediction of Rolling Element Bearings Using Supervised Machine Learning

Components of rotating machines, such as shafts, bearings and gears are subject to performance degradation, which if left unattended could lead to failure or breakdown of the entire system. Analyzing condition monitoring data, implementing diagnostic techniques and using machinery prognostic algorithms will bring about accurate estimation of the remaining life and possible failures that may occur. This paper proposes a combination of two supervised machine learning techniques; namely, the regression model and multilayer artificial neural network model, to predict the remaining useful life of rolling element bearings. Root mean square and Kurtosis were analyzed to define the bearing failure stages. The proposed methodology was validated through two case studies involving vibration measurements of an operational wind turbine gearbox and a split cylindrical roller bearing in a test rig.

Prognosis of a Wind Turbine Gearbox Bearing Using Supervised Machine Learning

Deployment of large-scale wind turbines requires sophisticated operation and maintenance strategies to ensure the devices are safe, profitable and cost-effective. Prognostics aims to predict the remaining useful life (RUL) of physical systems based on condition measurements. Analyzing condition monitoring data, implementing diagnostic techniques and using machinery prognostic algorithms will bring about accurate estimation of the remaining life and possible failures that may occur. This paper proposes to combine two supervised machine learning techniques, namely, regression model and multilayer artificial neural network model, to predict the RUL of an operational wind turbine gearbox using vibration measurements. Root Mean Square (RMS), Kurtosis (KU) and Energy Index (EI) were analysed to define the bearing failure stages. The proposed methodology was evaluated through a case study involving vibration measurements of a high-speed shaft bearing used in a wind turbine gearbox.