Estimated Success Rate Research Articles

A Supervised Learning Framework for Joint Estimation of Angles-of-Arrival and Number of Sources

Machine learning is a promising technique for angle-of-arrival (AOA) estimation of waves impinging a sensor array. However, the majority of the methods proposed so far only consider a known, fixed number of impinging waves, i.e., a fixed number of sources (NOS). This paper proposes a machine-learning-based estimator designed for the case when the NOS is variable and hence unknown a priori. The proposed estimator comprises a framework of single-label classifiers. Each classifier predicts if waves are present within certain randomly selected segments of the array’s field of view (FOV), resulting from discretising the FOV with a certain (FOV) resolution. The classifiers’ predictions are combined into a probabilistic angle spectrum, whereupon the NOS and the AOAs are estimated jointly by applying a probability threshold whose optimal level is learned from data. The estimator’s performance is assessed using a new performance metric: the joint AOA estimation success rate. Numerical simulations show that for low SNR (−10 dB), a low FOV resolution (2°) yields a higher success rate than a high resolution (1°), whereas the opposite applies for mid (0 dB) and high (10 dB) SNRs. In nearly all simulations, except one at low SNR and a high FOV resolution, the proposed estimator outperforms the MUSIC algorithm if the maximum allowed AOA estimation error is approximately equal to (or larger than) the FOV resolution.

Long-Term Outcomes of Bowman Layer Inlay Transplantation for the Treatment of Progressive Keratoconus.

To report long-term clinical outcomes and estimated success rates after Bowman layer (BL) inlay transplantation in eyes treated for progressive keratoconus (KC). Thirty-five eyes (29 patients) with progressive KC underwent BL inlay transplantation. Best-corrected spectacle and contact lens visual acuity, Scheimpflug-based corneal tomography [simulated and maximum keratometry (Kmax)], central corneal thickness, thinnest point thickness, complications, and success rate were evaluated up to 8 years postoperatively for the total group and 2 subgroups [group 1: preoperative Kmax > 69 diopter (D) (n = 26); group 2: preoperative Kmax < 69D (n = 9)]. Mean logarithm of the minimum angle of resolution best contact lens-corrected visual acuity for the total group and the 2 subgroups did not change from preoperative up to the last available follow-up (all P > 0.05), whereas best spectacle-corrected visual acuity improved for group 1 ( P = 0.03). Group 1 showed an average Kmax reduction of 7D in the first month ( P < 0.001) with no further changes up to 8 years postoperatively ( P > 0.05), whereas no significant changes were observed in group 2 (all P > 0.05). Postoperative KC progression occurred in 4 eyes (n = 3 group 1, n = 1 group 2) and 1 eye (group 2) underwent retransplantation for unsatisfactory visual performance. Kaplan-Meier analysis showed an estimated success rate of 85% at the 5 to 8 years follow-up in group 1 and of 75% at the 5 to 7 years follow-up for group 2. BL inlay transplantation stabilized KC in most eyes along with preservation of contact lens tolerance up to 8 years postoperatively and may be a successful treatment option, in particular for eyes with progressive advanced KC (Kmax > 69D).

Preferences and experiences regarding the treatment of obstructive sleep apnea with mandibular advancement splints - a cross-sectional pilot survey

ABSTRACT Objective To obtain a current overview of the use of mandibular advancement splints (MAS) as a therapeutic option for obstructive sleep apnea (OSA) in the dental profession in Northern Germany. Methods A questionnaire was mailed to 2431 dental professionals, and 8.3% responded. Descriptive data analysis was performed on each question separately. Results Most of the surveyed dentists (51.7%) estimated the success rate of MAS therapy to be higher than 75%. Analysis of the relationship between the number of splints prescribed annually and the dental specialization (p = 0.22), work experience (p = 0.14), estimated success rate of MAS therapy (p = 0.96), and discontinuation of MAS therapy (p = 0.57) revealed no significant correlations. Conclusion Mandibular advancement splints seem to be a safe and successful therapy for OSA. There seems to be a lack of information on the patient side and a lack of education on the dentist side.

185 Vascular Hitch Procedure for Accessory Lower Pole Vessels

Abstract Introduction The gold standard procedure for pelvi-ureteric junction obstruction has been the Anderson-Hynes dismembered pyeloplasty; involving the repositioning of the ureter and ureteropelvic anastomosis. However, the Hellstroem 'Vascular Hitch Procedure’ dictates the superior translocation of the accessory vessel and its fixation to the anterior pelvic wall. The latter has an estimated success rate &amp;gt;90%. Method During 2016-2020, at Leicester Royal Infirmary, 16 operations occurred on paediatric patients with pelvi-ureteric junction obstruction. The dismembered pyeloplasty was performed on 5 patients, 9 patients underwent the vascular hitch procedure, and 2 patients are currently awaiting the latter operation. All patients had a pre-operative functional magnetic resonance urography (fMRU) to identify and localise the accessory lower pole vessel. Results The mean hospital stay for the vascular hitch procedure was 1.5 days (range=1-2) in comparison to 4 days (range=3-5) for the dismembered pyeloplasty. The follow-up period ranged from 6 months to 3 years. Overall, it was noted that the patients were asymptomatic, had markedly reduced hydronephrosis on imaging as well as stable renal function noted on the MAG 3 renogram scan. Conclusions The laparoscopic vascular hitch procedure appears to be the superior operation for the management of pelvi-uteric junction obstruction as the patients had notably reduced hospital stay lengths.

Profitability analysis with the fuzzy logic: A hospital example

Implementation of the fuzzy logic is a modern approach for cost-volume-profit analysis and decision-making process under risk and uncertainty (Yuan, 2009). The implementation of the fuzzy logic approach especially makes sense for profit or loss estimations in developing countries, where uncertainties and risks are often observed (Roztocki &amp; Weistroffer, 2005). This study aimed to estimate the profit or loss of indirect Coombs blood test, which is among the 100 blood tests run by the laboratory department of a healthcare organization located in Istanbul, Turkey, that started operations in 2018. Another purpose of the research was to compare the profit or loss estimated by fuzzy logic with the actual values. Research questions of the study were: 1) Can fuzzy logic be used in the health sector’s profitability estimates? 2) What is the estimated success rate of fuzzy logic in the case of uncertainty and complexity? 3) If the fuzzy logic can be used in the health sector’s profit forecasts, how close are the estimated profit sums achieved by the fuzzy logic to the actual profit sums? Based on the findings of the study, profit estimated by the fuzzy logic is in a close range to actual values with a low error rate

Planted Forest Fire Burn Area and Impact Assessment Using Sentinel-2: Case Study of the University of Ilorin Teak Plantation

This paper presents a comprehensive assessment of the locations, extent and the impact of forest fire in University of Ilorin Teak Plantation using pre- and post-fire Sentinel-2 level 1C products. First, the pre-fire image was classified into three classes: vegetation area, bare soil and water body, using supervised classification (Maximum Likelihood method) to distinguish between vegetation and non-vegetation areas. Then, from the post-fire image, the burn areas were detected and extracted using Normalized Burnt Ratio. With the burn area polygon, impact of the fire on the planted forest was determined by isolating the vegetation class within the classified map so estimating the number of teak trees affected through extrapolation of the burn area and the tree spacing grid of 3m. The classification result shows that vegetation land cover type accounted for about 419.7 ha (66 %) of the total area while bare soil and water body take 204.3 ha (32 %) and 12.9 ha (2 %), respectively. Also, the resulting classified map produced overall classification accuracy of 95 %. Impact assessment result reveals that a total number of 49156 tree stands were affected by the fire within burnt area of 54.8 ha (8.6%). Analysis of the estimation success rate using one of the burn areas as validation site yielded approximation in excess of 3% with 17621 counted and 18222 estimated. Planted forest management and planning has many phases; so, it is necessary to understand the current and future condition of what is being manage. The fire burn map derived from this study will assist the University teak plantation management team update its current management strategy to protect it from continuous exposure to fire. From fire management perspective, the list of planning activities that require future assessments include pruning preferences, replanting, commercial thinning, spacing of planted trees, and perimeter buffering.

6D Pose Estimation of Occlusion-Free Objects for Robotic Bin-Picking Using PPF-MEAM With 2D Images (Occlusion-Free PPF-MEAM)

Pose estimation that locates objects in a bin is necessary for a robotic bin picking system. Although many algorithms have shown high performance in pose estimation, most algorithms estimate the poses of objects regardless of their occlusion. This can reduce the success rate in picking up the object. To resolve this issue, we propose a novel pipeline that estimates a pose only for occlusion-free objects based on point pair feature-based pose estimation with multiple edge appearance model (PPF-MEAM). The proposed method detects occlusion-free objects in the 2D image captured by a camera with a convolutional neural network framework. Next, corresponding point clouds of occlusion-free objects need to be extracted by using their locations in the 2D image. we propose a robust extraction method that finds the 3D points corresponding to image pixels in the 2D image to reduce the effect of the calibration errors between the camera and 3D sensor. The point cloud of the occlusion-free objects is finally input into a pipeline of PPF-MEAM to estimate the pose of the object. The experiment results prove that the proposed method is about 50% faster 30% higher in terms of pose estimation success rate compared with the original PPF. Moreover, it increases the success rate of picking tasks compared with the original PPF-MEAM.

Can the detection dog alert on COVID-19 positive persons by sniffing axillary sweat samples? A proof-of-concept study.

The aim of this proof-of-concept study was to evaluate if trained dogs could discriminate between sweat samples from symptomatic COVID-19 positive individuals (SARS-CoV-2 PCR positive) and those from asymptomatic COVID-19 negative individuals. The study was conducted at 2 sites (Paris, France, and Beirut, Lebanon), followed the same training and testing protocols, and involved six detection dogs (three explosive detection dogs, one search and rescue dog, and two colon cancer detection dogs). A total of 177 individuals were recruited for the study (95 symptomatic COVID-19 positive and 82 asymptomatic COVID-19 negative individuals) from five hospitals, and one underarm sweat sample per individual was collected. The dog training sessions lasted between one and three weeks. Once trained, the dog had to mark the COVID-19 positive sample randomly placed behind one of three or four olfactory cones (the other cones contained at least one COVID-19 negative sample and between zero and two mocks). During the testing session, a COVID-19 positive sample could be used up to a maximum of three times for one dog. The dog and its handler were both blinded to the COVID-positive sample location. The success rate per dog (i.e., the number of correct indications divided by the number of trials) ranged from 76% to 100%. The lower bound of the 95% confidence interval of the estimated success rate was most of the time higher than the success rate obtained by chance after removing the number of mocks from calculations. These results provide some evidence that detection dogs may be able to discriminate between sweat samples from symptomatic COVID-19 individuals and those from asymptomatic COVID-19 negative individuals. However, due to the limitations of this proof-of-concept study (including using some COVID-19 samples more than once and potential confounding biases), these results must be confirmed in validation studies.

An automatic method of fish length estimation using underwater stereo system based on LabVIEW

The underwater non-contact methods for fish length measurement play an important role in aquaculture for fish sorting, grading and effective management. One challenge is to obtain the fish length data automatically instead of to locate the key measurement points of fish manually during the processing. This study presents a stereo vision system with procedure that obtains the oplegnathus punctatus fish body length based on LabVIEW. The key measurement points of fish snout tip and caudal peduncle can be automatically extracted through the processing algorithm that is based on the contour and convex hull calculation. The experiment validated the success rate and accuracy of fish length estimation at different angles of the free-swimming fish body to the optical axis. Experimental results showed that the proposed system could estimate the fish length with high accuracy and success rate. The regression analysis between estimated and manually measured body length showed a good linear relationship with R2 > 0.91 and mean relative error <2.55%. These findings indicated that the proposed system for non-contact and automatic fish body length measurement could be applied in the fish farm.

Vemurafenib in non-small-cell lung cancer patients with BRAFV600 and BRAFnonV600 mutations

BRAF mutations occurring in 1%-5% of patients with non-small-cell lung cancer (NSCLC) are therapeutic targets for these cancers but the impact of the exact mutation on clinical activity is unclear. The French National Cancer Institute (INCA) launched the AcSé vemurafenib trial to assess the efficacy and safety of vemurafenib in cancers with various BRAF mutations. We herein report the results of the NSCLC cohort. Tumour samples were screened for BRAF mutations in INCA-certified molecular genetic centres. Patients with BRAF-mutated tumours progressing after ≥1 line of treatment were proposed vemurafenib 960 mg twice daily. Between October 2014 and July 2018, 118 patients were enrolled in the NSCLC cohort. The primary outcome was the objective response rate (ORR) assessed every 8 weeks (RECIST v1.1). A sequential Bayesian approach was planned with an inefficacy bound of 10% for ORR. If no early stopping occurred, the treatment was of interest if the estimated ORR was ≥30% with a 90% probability. Secondary outcomes were tolerance, response duration, progression-free survival (PFS), and overall survival (OS). Of the 118 patients enrolled, 101 presented with a BRAFV600 mutation and 17 with BRAFnonV600 mutations; the median follow-up was 23.9 months. In the BRAFnonV600 cohort, no objective response was observed and this cohort was stopped. In the BRAFV600 cohort, 43/96 patients had objective responses. The mean Bayesian estimated success rate was 44.9% [95% confidence intervals (CI) 35.2%-54.8%]. The ORR had a 99.9% probability of being ≥30%. Median response duration was 6.4 months, median PFS was 5.2 months (95% CI 3.8-6.8), and OS was 10 months (95% CI 6.8-15.7). The vemurafenib safety profile was consistent with previous publications. Routine biomarker screening of NSCLC should include BRAFV600 mutations. Vemurafenib monotherapy is effective for treating patients with BRAFV600-mutated NSCLC but not those with BRAFnonV600 mutations. ClinicalTrials.gov identifier: NCT02304809.

Timing the Provision of Product Information in Reward-Based Crowdfunding

In a crowdfunding campaign, customers are often uncertain about how much they like the product launched because the product description often lacks detailed information, in which case the product value for a customer is her expected valuation. When that information is provided, customers' valuations are realized and become more heterogeneous. We study a dynamic model of reward-based crowdfunding in which customers' arrivals follow a Poisson process. Upon arrival, a customer's pledging decision depends on her valuation of the product, which is determined by the product information observed at that time, and the estimated success rate of the campaign, which is modeled by using the rational expectation equilibrium framework. The creator faces a problem of deciding when to disclose the detailed information about the product to the customers during campaign. We characterize the optimal timing of disclosing detailed information for the creator if his objective is to maximize the success rate of the campaign. We find that it is optimal to disclose detailed product information at the very beginning if the pledging cost is sufficiently high or the market size is sufficiently small, and it is optimal to delay the disclosure otherwise. We also show that increasing the customers heterogeneity at a right time can benefit the creator. This result makes our work different from traditional advance selling theory (Shugan and Xie (2000)) which states that advance selling can benefit a seller by homogenizing heterogeneous buyers.

Extended long-short ambiguity resolution in multi-antenna GNSS-over-fiber systems for enhanced attitude determination

An extended long-short ambiguity resolution (AR) approach to multi-antenna GNSS-over-fiber systems is proposed and experimentally demonstrated, providing enhanced attitude determination. GNSS signals are remotely acquired from M extended antennas and then transferred to a central node. Through virtual baseline processing, multiple baselines with intermediate length are constructed. Consequently, although the adjacent baseline ratio is limited, the overall aperture is expanded. The carrier phase measurements of equal length virtual baselines are averaged by following the parity detection process for reducing noise, leading to refined baseline estimation. Also, the extended long-short AR method is a single-epoch and search-free one, enabling both improved accuracy and real-time implementations. Both the AR success rate and accuracy of baseline estimation are remarkably enhanced, thus resulting in enhanced attitude determination. Full 3-D attitude determination is also possible with the proposed angle-type or cross-type planar array. In the experiments, the ambiguity is successfully resolved for more than 1400 epochs. The standard deviation of the baseline vector is reduced to almost one-third in both horizontal and vertical directions, without specific delay monitoring hardware.

Multiple Preambles for High Success Rate of Grant-Free Random Access With Massive MIMO

Grant-free random access (RA) with massive MIMO is a promising RA technique with low signaling overhead that provides significant benefits in increasing the channel reuse efficiency. Since user equipment (UE) detection and channel estimation in grant-free RA rely solely on the received preambles, preamble designs that enable high success rate of UE detection and channel estimation are very much in need to ensure the performance gain of grant-free RA with massive MIMO. In this paper, a super preamble consisting of multiple consecutive preambles is proposed for high success rate of grant-free RA with massive MIMO. With the proposed approach, the success of UE detection and channel estimation for a RA UE depends on two conditions: 1) it is a solvable UE; 2) its super preamble is detected. Accordingly, we theoretically analyze the solvable rate of RA UEs with multiple preambles and propose a reliable UE detection algorithm to obtain the super preambles of RA UEs by exploiting the quasi-orthogonality characteristic of massive MIMO. Theoretical analysis and simulation results show that turning a preamble into a super preamble consisting of two or three shorter preambles, the success rate of UE detection and channel estimation could be significantly increased using the proposed approach.

Assessing the Protective Metabolome Using Machine Learning in World Trade Center Particulate Exposed Firefighters at Risk for Lung Injury

The metabolome of World Trade Center (WTC) particulate matter (PM) exposure has yet to be fully defined and may yield information that will further define bioactive pathways relevant to lung injury. A subset of Fire Department of New York firefighters demonstrated resistance to subsequent loss of lung function. We intend to characterize the metabolome of never smoking WTC-exposed firefighters, stratified by resistance to WTC-Lung Injury (WTC-LI) to determine metabolite pathways significant in subjects resistant to the loss of lung function. The global serum metabolome was determined in those resistant to WTC-LI and controls (n = 15 in each). Metabolites most important to class separation (top 5% by Random Forest (RF) of 594 qualified metabolites) included elevated amino acid and long-chain fatty acid metabolites, and reduced hexose monophosphate shunt metabolites in the resistant cohort. RF using the refined metabolic profile was able to classify cases and controls with an estimated success rate of 93.3%, and performed similarly upon cross-validation. Agglomerative hierarchical clustering identified potential influential pathways of resistance to the development of WTC-LI. These pathways represent potential therapeutic targets and warrant further research.

Sex estimation using external morphology of the frontal bone and frontal sinuses in a contemporary Czech population.

Sex estimation is a task of utmost importance in forensic anthropology and bioarcheology. Along with the pelvic bone, the skull is the most important source of sexual dimorphism. On the human skull, the upper third of the face (i.e., the frontal bone) is one of the most significant sexually dimorphic structures useful in anthropological research, especially when studied by methods of virtual anthropology. This study was focused on sex estimation using the form and shape of the external surface of the frontal bone with or without the inclusion of its sinuses. The study sample consisted of 103 cranial CT images from a contemporary Czech population. Three-dimensional virtual models of the frontal bones and sinuses were analyzed using geometric morphometrics and multidimensional statistics: coherent point drift-dense correspondence analysis (CPD-DCA), principal component analysis (PCA), and support vector machine (SVM). The whole external frontal surface was significantly different between males and females both in form and shape. The greatest total success rate of sex estimation based on form was 93.2%, which decreased to 86.41% after crossvalidation, and this model identified females and males with the same accuracy. The best estimation based on shape reached a success rate of 91.26%, with slightly greater accuracy for females. After crossvalidation, however, the success rate decreased to 83.49%. The differences between sexes were significant also in the volume and surface of the frontal sinuses, but the sex estimation had only 64.07% accuracy after crossvalidation. Simultaneous use of the shape of the frontal surface and the frontal sinuses improved the total success rate to 98.05%, which decreased to 84.46% after crossvalidation.

Real-time Ultrasound-Guided Lumbar Epidural with Transverse Interlaminar View: Evaluation of an In-Plane Technique.

The anatomical landmarks method is currently the most widely used technique for epidural needle insertion and is faced with multiple difficulties in certain patient populations. Real-time ultrasound guidance has been recently used to aid in epidural needle insertion, with promising results. Our aim was to test the feasibility, success rate, and satisfaction associated with a novel real-time ultrasound-guided lumbar epidural needle insertion in the transverse interlaminar view. Prospective descriptive trial on a novel approach. Operating room and preoperative holding area at a tertiary care hospital. Adult patients presenting for elective open prostatectomy and planned for surgical epidural anesthesia. Consented adult patients aged 30-80 years scheduled for open prostatectomy under epidural anesthesia were enrolled. Exclusion criteria included allergy to local anesthetics, infection at the needle insertion site, coagulopathy, and patient refusal. A curvilinear low-frequency (2-5 MHz) ultrasound probe and echogenic 17-G Tuohy needles were used by one of three attending anesthesiologists. Feasibility of epidural insertion was defined as a 90% success rate within 10 minutes. Twenty-two patients were enrolled into the trial, 14 (63.6%) of whom found the process to be satisfactory or very satisfactory. The median time to perform the block was around 4.5 minutes, with an estimated success rate of 95%. No complications related to the epidural block were observed over the 48 hours after the procedure. We demonstrate the feasibility of a novel real-time ultrasound-guided epidural with transverse interlaminar view.

Reducing the Number of Tests Cases for Reliability Assessment of Protective Relays

This paper presents a methodology to perform application tests in protective relays. It is based on using the Gabriel graph to select, from a large number of events generated by a Monte Carlo-based procedure, a smaller subset that is able to predict the relay performance without compromising the quality of the estimator. The proposed methodology is successfully applied to test the behavior of differential protective relays in two situations: turn-to-ground faults and transformer energizations. In both cases, the proposed methodology significantly reduces the number of tests required to predict the success rate of the tested relays for a desired confidence level and variance. It is observed that the efficiency of the method increases as the estimated success rate of the relay increases. This means that if the estimated success rate is high, the proposed methodology leads to larger reduction in the number of events that are actually necessary to test the relay. The validity of the proposed methodology is demonstrated by means of tests performed with an actual protective relay.

Estimating Vehicle Movement Direction from Smartphone Accelerometers Using Deep Neural Networks.

Characterization of driving maneuvers or driving styles through motion sensors has become a field of great interest. Before now, this characterization used to be carried out with signals coming from extra equipment installed inside the vehicle, such as On-Board Diagnostic (OBD) devices or sensors in pedals. Nowadays, with the evolution and scope of smartphones, these have become the devices for recording mobile signals in many driving characterization applications. Normally multiple available sensors are used, such as accelerometers, gyroscopes, magnetometers or the Global Positioning System (GPS). However, using sensors such as GPS increase significantly battery consumption and, additionally, many current phones do not include gyroscopes. Therefore, we propose the characterization of driving style through only the use of smartphone accelerometers. We propose a deep neural network (DNN) architecture that combines convolutional and recurrent networks to estimate the vehicle movement direction (VMD), which is the forward movement directional vector captured in a phone’s coordinates. Once VMD is obtained, multiple applications such as characterizing driving styles or detecting dangerous events can be developed. In the development of the proposed DNN architecture, two different methods are compared. The first one is based on the detection and classification of significant acceleration driving forces, while the second one relies on longitudinal and transversal signals derived from the raw accelerometers. The final success rate of VMD estimation for the best method is of 90.07%.

Should we expand the indications for the DAIR (debridement, antibiotic therapy, and implant retention) procedure for Staphylococcus aureus prosthetic joint infections? A multicenter retrospective study.

To evaluate factors associated with failure in patients treated with DAIR (debridement, antibiotic therapy, and implant retention) for Staphylococcus aureus prosthetic joint infections (PJIs). We retrospectively analyzed consecutive patients with stable PJI due to S. aureus treated with DAIR at six hospitals between 2010 and 2014. Cox proportional hazards regression was used to study factors associated with treatment failure at 2years. Of 154 eligible patients, 137 were included (mean age 73 ± 13 years; male 56%). The estimated success rate according to the Kaplan-Meier method was 76.2 [95% CI 68-83] at 2years of follow-up. In multivariate analysis, longer duration of treatment (hazard ratio (HR) 0.78 [0.69-0.88]; p < 0.001) and combination therapy including rifampin (HR 0.08 [0.018-0.36]; p = 0.001) were independently associated with success, whereas active smoking was independently associated with failure (HR 3.6 [1.09-11.84]; p = 0.036). When the analysis was restricted to patients with early infection onset (< 3months), early acute infection was also predictive of a better prognosis (HR 0.25 [0.09-0.7]; p = 0.009). Failure was not associated with time from prosthesis insertion to debridement, nor with duration of symptoms > 3weeks and type of prosthesis (hip or knee). These results remained unchanged when the 14 patients under immunosuppressive therapy were removed from analysis. These data suggest that DAIR can be performed even if infection and symptoms are delayed but reserved to patients who are able to follow rifampin-based combination therapy for a prolonged duration that should not be different for hip and knee PJI.

Metabolomics of Protection from the Development of World Trade Center-Lung Injury: A Machine Learning Approach

INTRODUCTION: The metabolome remains an untapped resource with potential to comprehensively characterize and identify bioactive pathways of disease after WTC-PM exposure. A subset of rescue/recovery workers from the FDNY demonstrated resistance to subsequent loss of lung function. We have previously identified biomarkers significantly associated with protection from WTC-LI. OBJECTIVE: Implement machine learning and dimension reduction techniques to characterize the metabolome of WTC-exposed firefighters, stratified by resistance to WTC-LI, to identify metabolite pathways of significance to resistance to loss of lung disease post PM exposure. METHODS: Serum global metabolome of never-smoker, WTC-exposed firefighters with normal pre-9/11 lung function, segregated by forced expiratory volume-one second(FEV1) at symptomatic presentation. Control (n=15) from previously identified cohort-control and resistant to WTC-LI (FEV1, %- Pred≥107; within one-standard deviation of the highest FEV1% predicted of the study cohort, n=15) were selected. Pathways/metabolites associated with resistance to WTC-LI were determined using dimension reduction techniques. RESULTS: Metabolites most important to class separation (top 5% by RF of 594 qualified metabolites), included elevated amino acid and long-chain fatty acid metabolites, and reduced hexose monophosphate shunt metabolites in the resistant cohort. RF of the refined metabolic profile correctly classified cases and controls with a 93.3% estimated success rate. Agglomerative hierarchical clustering identified potential mechanistic relations. CONCLUSION: High-dimensional data analysis on the metabolome of a subset of WTC-PM exposed 9/11 rescue workers has identified pathways associated with resistance to loss of lung function after an acute PM exposure. These pathways represent novel potential therapeutic targets and warrant further research. FUNDING STATEMENT: NHLBI R01HL119326, CDC/NIOSH U01-OH11300, Clinical Center of Excellence 200-2017-93426, Data Center 200-2017-93326. This work was also partially funded by the NYU-HHC CTSI supported by grant UL1T38 from the National Center for Advancing Translational Sciences of the NIH and by the Saperstein Scholars Fund. DECLARATION OF INTERESTS: The authors report no conflicts of interest. ETHICS APPROVAL STATEMENT: Subjects, at the time of enrollment, consented to analysis of their information and samples for research according to Institutional Review Board (IRB) approved protocols at Montefiore Medical Center(#07-09-320) and New York University(#16-01412). All experiments conform to the relevant regulatory standards as per IRB standards.