An Automated Line-of-Therapy Algorithm for Adults With Metastatic Non-Small Cell Lung Cancer: Validation Study Using Blinded Manual Chart Review.

BACKGROUND: Extraction of line-of-therapy (LOT) information from electronic health record and claims data is essential for determining longitudinal changes in systemic anticancer therapy in real-world clinical settings. OBJECTIVE: The aim of this retrospective cohort analysis is to validate and refine our previously described open-source LOT algorithm by comparing the output of the algorithm with results obtained through blinded manual chart review. METHODS: We used structured electronic health record data and clinical documents to identify 500 adult patients treated for metastatic non-small cell lung cancer with systemic anticancer therapy from 2011 to mid-2018; we assigned patients to training (n=350) and test (n=150) cohorts, randomly divided proportional to the overall ratio of simple:complex cases (n=254:246). Simple cases were patients who received one LOT and no maintenance therapy; complex cases were patients who received more than one LOT and/or maintenance therapy. Algorithmic changes were performed using the training cohort data, after which the refined algorithm was evaluated against the test cohort. RESULTS: For simple cases, 16 instances of discordance between the LOT algorithm and chart review prerefinement were reduced to 8 instances postrefinement; in the test cohort, there was no discordance between algorithm and chart review. For complex cases, algorithm refinement reduced the discordance from 68 to 62 instances, with 37 instances in the test cohort. The percentage agreement between LOT algorithm output and chart review for patients who received one LOT was 89% prerefinement, 93% postrefinement, and 93% for the test cohort, whereas the likelihood of precise matching between algorithm output and chart review decreased with an increasing number of unique regimens. Several areas of discordance that arose from differing definitions of LOTs and maintenance therapy could not be objectively resolved because of a lack of precise definitions in the medical literature. CONCLUSIONS: Our findings identify common sources of discordance between the LOT algorithm and clinician documentation, providing the possibility of targeted algorithm refinement.

Deep learning for classification of pediatric chest radiographs by WHO's standardized methodology.

BACKGROUND: The World Health Organization (WHO)-defined radiological pneumonia is a preferred endpoint in pneumococcal vaccine efficacy and effectiveness studies in children. Automating the WHO methodology may support more widespread application of this endpoint. METHODS: We trained a deep learning model to classify pneumonia CXRs in children using the World Health Organization (WHO)'s standardized methodology. The model was pretrained on CheXpert, a dataset containing 224,316 adult CXRs, and fine-tuned on PERCH, a pediatric dataset containing 4,172 CXRs. The model was then tested on two pediatric CXR datasets released by WHO. We also compared the model's performance to that of radiologists and pediatricians. RESULTS: The average area under the receiver operating characteristic curve (AUC) for primary endpoint pneumonia (PEP) across 10-fold validation of PERCH images was 0.928; average AUC after testing on WHO images was 0.977. The model's classification performance was better on test images with high inter-observer agreement; however, the model still outperformed human assessments in AUC and precision-recall spaces on low agreement images. CONCLUSION: A deep learning model can classify pneumonia CXR images in children at a performance comparable to human readers. Our method lays a strong foundation for the potential inclusion of computer-aided readings of pediatric CXRs in vaccine trials and epidemiology studies.

Pneumococcal vaccination coverage among adults aged 19 to 64 years with immuno-compromising conditions, cerebrospinal fluid (CSF) leaks, or cochlear implants in the US.

Background: Adults with immuno-compromising conditions, CSF leaks, or cochlear implants are at increased risk for pneumococcal disease (high-risk patients), yet pneumococcal vaccination rates in the US for this group are low.Methods: A retrospective cohort analysis was conducted from 2010 to 2018 using the Truven Health MarketScan database to estimate pneumococcal vaccination coverage among adults aged 19 to 64 years newly diagnosed with high-risk conditions, and to assess factors associated with receiving the recommended pneumococcal vaccines.Results: The study sample included 2,497,799 adults aged 19 to 64 years old with newly diagnosed high-risk conditions. Most of the study cohort had seven or more annual physician office (52%) and pharmacy (56%) visits. The proportion of high-risk adults who received at least one pneumococcal vaccination increased from 5.4% after 1 year of follow-up to 14.2% after 6 years of follow-up. Compared to those who received no pneumococcal vaccination, high-risk adults who received any pneumococcal vaccination were more likely to be older, female, enrolled in an HMO, had more healthcare encounters, and were treated by a primary care provider.Conclusion: Despite numerous healthcare encounters annually, very few high-risk adults received pneumococcal vaccines, highlighting the need for implementing targeted interventions to increase vaccine uptake in this vulnerable population.

Outcomes of retreatment with anti-PD-1 monotherapy after response to first course in patients with cutaneous melanoma.

Aim: To determine outcomes of retreatment with anti-PD-1 monotherapy for melanoma. Materials & methods: This retrospective study included adults with unresectable cutaneous melanoma who achieved stable disease (SD) or better after anti-PD-1 monotherapy and were retreated with anti-PD-1 monotherapy after ≥90-day gap. We determined overall survival and real-world tumor response. Results: For 21 eligible patients, from retreatment initiation, median follow-up was 14.4 months (range, 2.6-34.5); median overall survival was 30.0 months (95% CI: 14.4-not reached); 1-year survival was 100% (95% CI: 100-100%); 2-year survival was 83% (48-96%). Of 16 patients with recorded best real-world tumor response, ten (63%) responded (complete/partial response); three achieved SD; three had progressive disease. Conclusion: Patients with advanced melanoma achieving SD/better after first-course anti-PD-1 monotherapy may benefit from retreatment.

The Impact of Varicella Vaccination on the Incidence of Varicella and Herpes Zoster in the United States: Updated Evidence From Observational Databases, 1991-2016.

BACKGROUND: Universal childhood vaccination against varicella began in the United States as a 1-dose schedule in 1996, changing to a 2-dose schedule in 2006. The exogenous boosting hypothesis, which postulates that reexposure to circulating wild-type varicella delays the onset of herpes zoster, predicts a transient increase in the incidence of herpes zoster, peaking in adults 15-35 years after the start of varicella vaccination. METHODS: This was a retrospective study of administrative claims data from the MarketScan Commercial and Medicare databases between 1991-2016. Outcome measures were the incidences of herpes zoster per 100 000 person-years, by calendar year and age category, and the annual rates of change in herpes zoster by age category, in an interrupted time series regression analysis, for the periods of 1991-1995 (prevaccine), 1996-2006 (1-dose vaccination period), and 2007-2016 (2-dose vaccination period). RESULTS: The annual incidences of herpes zoster increased throughout the period of 1991-2012 in all adult age categories, with a plateau in 2013-2016 that was most evident in the ≥65 age group. In 1991-1995, the herpes zoster incidences increased at annual rates of 4-6% in age categories 18-34, 35-44, 45-54, and 55-64 years. In the same age categories during 1996-2006 and 2007-2016, the herpes zoster incidences increased at annual rates of 1-5%. CONCLUSIONS: Although the annual incidence of herpes zoster in adults has continued to increase, the rates of change decreased during both the 1- and 2-dose vaccination periods. The hypothesized increase in herpes zoster predicted from modelling of the exogenous boosting hypothesis was not observed.

Temporal phenotyping by mining healthcare data to derive lines of therapy for cancer.

Lines of therapy (LOT) derived from real-world healthcare data not only depict real-world cancer treatment sequences, but also help define patient phenotypes along the course of disease progression and therapeutic interventions. The sequence of prescribed anticancer therapies can be defined as temporal phenotyping resulting from changes in morphological (tumor staging), biochemical (biomarker testing), physiological (disease progression), and behavioral (physician prescribing and patient adherence) parameters. We introduce a novel methodology that is a two-part approach: 1) create an algorithm to derive patient-level LOT and 2) aggregate LOT information via clustering to derive temporal phenotypes, in conjunction with visualization techniques, within a large insurance claims dataset. We demonstrated the methodology using two examples: metastatic non-small cell lung cancer and metastatic melanoma. First, we generated a longitudinal patient cohort for each cancer type and applied a set of rules to derive patient-level LOT. Then the LOT algorithm outputs for each cancer type were visualized using Sankey plots and K-means clusters based on durations of LOT and of gaps in therapy between LOT. We found differential distribution of temporal phenotypes across clusters. Our approach to identify temporal patient phenotypes can increase the quality and utility of analyses conducted using claims datasets, with the potential for application to multiple oncology disease areas across diverse healthcare data sources. The understanding of LOT as defining patients' temporal phenotypes can contribute to continuous health learning of disease progression and its interaction with different treatment pathways; in addition, this understanding can provide new insights that can be applied by tailoring treatment sequences for the patient phenotypes who will benefit.

Real-world experience with pembrolizumab in patients with advanced melanoma: A large retrospective observational study.

Pembrolizumab has been approved in the United States for treating advanced melanoma for >4 years. We examined real-world pembrolizumab use and associated outcomes in US oncology clinical practices, including patients who would not be eligible for clinical trials.Flatiron Health longitudinal database was used to identify adult patients with advanced melanoma initiating ≥1 dose of pembrolizumab from September 4, 2014, through December 31, 2016, with follow-up through December 31, 2017. Patients in any clinical trial during the study period were excluded. Overall survival (OS) and time on treatment from pembrolizumab initiation were analyzed using the Kaplan-Meier (KM) method. Subgroup analyses were conducted to examine OS for several patient characteristics including Eastern Cooperative Oncology Group (ECOG) performance status >1, brain metastases, and corticosteroids before pembrolizumab initiation.Pembrolizumab was administered to 315 (59%), 152 (29%), and 65 (12%) patients as first-, second-, and third-line/later therapy. Median age at pembrolizumab initiation was 68 years (range, 18-84); most patients were male (66%) and white (94%). Of those with available data, 38% had BRAF-mutant melanoma, 21% had elevated lactate dehydrogenase (LDH) level, and 23% had ECOG >1. Overall, 18% had brain metastases, and 23% were prescribed corticosteroids <3 months before initiating pembrolizumab. Median study follow-up was 12.9 months (range, 0.03-39.6). Median OS was 21.8 months (95% confidence interval [CI] 16.8-29.1); KM 1-year and 2-year survival rates were 61% and 48%, respectively; and median time on pembrolizumab treatment was 4.9 months (95% CI 3.7-5.5). Median OS for first-line pembrolizumab was not reached, and for second-line and third-line/later was 13.9 and 12.5 months, respectively (log-rank P = .0095). Significantly better OS (all P ≤.0014, log-rank test) was evident for patients with ECOG performance status (PS) of 0 to 1 (vs >1), normal (vs elevated) LDH level, and no (vs yes) corticosteroid prescription <3 months before. No difference was recorded in OS by brain metastases (log-rank P = .22) or BRAF mutation status (log-rank P = .90).These findings support effectiveness of pembrolizumab in the real-world clinical setting and provide important insights into patient characteristics and outcomes associated with pembrolizumab therapy for a heterogeneous patient population with advanced melanoma, including patients who would not be eligible for clinical trials.

Pneumococcal vaccination patterns among persons aged 65 years or older in the United States: A retrospective database analysis.

OBJECTIVE: This study investigated the patterns of pneumococcal disease vaccination, the time between two different pneumococcal vaccine doses and factors associated with series completion. METHODS: A retrospective claims database analysis was conducted using the Clinformatics DataMart™ database. Adults who turned 65 years between January 1st, 2013 to June 30th, 2017 and were continuously enrolled (≥15 months) in the Medicare Advantage plans to June 30th, 2017 were included in this study. Pneumococcal vaccination patterns included: PCV13-PPV23, PPV23-PCV13, or receiving PPV23 or PCV13 only. Pneumococcal vaccination series completion was defined as receiving PCV13-PPV23 or PPV23-PCV13 from 65 years old to June 30th, 2017 while non-completion was defined as receiving only PCV13 or only PPV23 from 65 years old to June 30th, 2017. A multivariable logistic regression model was used to identify factors associated with pneumococcal vaccination series completion. RESULTS: A total of 224,132 adults were included in this study. Most received no pneumococcal vaccination (49%), while 34.3% received only one vaccine. Series completion occurred in 16.8% of adults. Some adults received only one vaccination: 11.6% received PPV23 and 22.7% received PCV13. The mean time between vaccinations was 420.8 days (approximately 14 months) for the PCV-PPV23 series, and 595.5 days (approximately 20 months) for the PPV23-PCV13 series. Adults were significantly more likely to complete pneumococcal vaccination series if they had at least one doctor's office, outpatient visit, or pharmacy visit versus no visits, or received an influenza vaccination in the first year after turning 65 years than those who did not (All: P < 0.001). CONCLUSION: Despite the 2014 recommendation, percentages of pneumococcal vaccination series completion were found to be low, aligning with recent literature. This highlights the need to improve series completion, given the increased risk and associated economic burden of pneumococcal disease in adults aged ≥65 years.

The real informatics challenges of personalized medicine: not just about the number of central processing units.

Personalized medicine has the potential to revolutionize patient care. In order to do so, it requires a re-engineering of many life sciences and healthcare processes, the most significant being the integration of the complete biomarker lifecycle from discovery to targeted treatment of patients. Individual patient omic profiles have become a reality owing to the diminishing cost of DNA sequencing. However, managing these data has created a bottleneck due to: the limitations in storage, computing power and information access; the lack of biologist-friendly software to replace the user-unfriendly custom scripts, which are crippling collaboration; the urgency for standardizing data across omics and clinical data realms for cross-study comparisons; undermining innovations of enterprise and open-source software, which saps innovations of open-source and reliability and support of enterprise software; and unavailability of a robust, integrative workflow system, which leads to actionable data at the patient care level.

Combining spatial priors and anatomical information for fMRI detection.

In this paper, we analyze Markov Random Field (MRF) as a spatial regularizer in fMRI detection. The low signal-to-noise ratio (SNR) in fMRI images presents a serious challenge for detection algorithms, making regularization necessary to achieve good detection accuracy. Gaussian smoothing, traditionally employed to boost SNR, often produces over-smoothed activation maps. Recently, the use of MRF priors has been suggested as an alternative regularization approach. However, solving for an optimal configuration of the MRF is NP-hard in general. In this work, we investigate fast inference algorithms based on the Mean Field approximation in application to MRF priors for fMRI detection. Furthermore, we propose a novel way to incorporate anatomical information into the MRF-based detection framework and into the traditional smoothing methods. Intuitively speaking, the anatomical evidence increases the likelihood of activation in the gray matter and improves spatial coherency of the resulting activation maps within each tissue type. Validation using the receiver operating characteristic (ROC) analysis and the confusion matrix analysis on simulated data illustrates substantial improvement in detection accuracy using the anatomically guided MRF spatial regularizer. We further demonstrate the potential benefits of the proposed method in real fMRI signals of reduced length. The anatomically guided MRF regularizer enables significant reduction of the scan length while maintaining the quality of the resulting activation maps.

Multimodal functional imaging using fMRI-informed regional EEG/MEG source estimation.

We propose a novel method, fMRI-Informed Regional Estimation (FIRE), which utilizes information from fMRI in E/MEG source reconstruction. FIRE takes advantage of the spatial alignment between the neural and the vascular activities, while allowing for substantial differences in their dynamics. Furthermore, with a region-based approach, FIRE estimates the model parameters for each region independently. Hence, it can be efficiently applied on a dense grid of source locations. The optimization procedure at the core of FIRE is related to the re-weighted minimum-norm algorithms. The weights in the proposed approach are computed from both the current source estimates and fMRI data, leading to robust estimates in the presence of silent sources in either fMRI or E/MEG measurements. We employ a Monte Carlo evaluation procedure to compare the proposed method to several other joint E/MEG-fMRI algorithms. Our results show that FIRE provides the best trade-off in estimation accuracy between the spatial and the temporal accuracy. Analysis using human E/MEG-fMRI data reveals that FIRE significantly reduces the ambiguities in source localization present in the minimum-norm estimates, and that it accurately captures activation timing in adjacent functional regions.

Multimodal functional imaging using fMRI-informed regional EEG/MEG source estimation.

We propose a novel method, fMRI-Informed Regional Estimation (FIRE), which utilizes information from fMRI in E/MEG source reconstruction. FIRE takes advantage of the spatial alignment between the neural and the vascular activities, while allowing for substantial differences in their dynamics. Furthermore, with the regional approach, FIRE can be efficiently applied to a dense grid of sources. Inspection of our optimization procedure reveals that FIRE is related to the re-weighted minimum-norm algorithms, the difference being that the weights in the proposed approach are computed from both the current estimates and fMRI data. Analysis of both simulated and human fMRI-MEG data shows that FIRE reduces the ambiguities in source localization present in the minimum-norm estimates. Comparisons with several joint fMRI-E/MEG algorithms demonstrate robustness of FIRE in the presence of sources silent to either fMRI or E/MEG measurements.

Study of neurovascular coupling in humans via simultaneous magnetoencephalography and diffuse optical imaging acquisition.

By combining diffuse optical imaging (DOI) and magnetoencephalography (MEG) we investigate neurovascular coupling non-invasively in human subjects using median-nerve stimulation. Previous fMRI studies have shown a habituation effect in the hemodynamic blood oxygen level-dependent (BOLD) response for stimulation periods longer than 2 s. With DOI and MEG we can test whether this effect in hemodynamic response can be accounted for by a habituation effect in the neural response. Our experimental results show that the habituation effect in the hemodynamic response is stronger than that in the earliest cortical neural response (N20). Using a linear convolution model to predict hemodynamic responses we found that including late neural components (> or = 30 ms) improves the prediction of the hemoglobin response. This finding suggests that in addition to the initial evoked-response deflections related to the talamic afferent input, later cortical activity is needed to predict the hemodynamic response.

A distributed spatio-temporal EEG/MEG inverse solver.

We propose a novel l(1)l(2)-norm inverse solver for estimating the sources of EEG/MEG signals. Based on the standard l(1)-norm inverse solvers, this sparse distributed inverse solver integrates the l(1)-norm spatial model with a temporal model of the source signals in order to avoid unstable activation patterns and "spiky" reconstructed signals often produced by the currently used sparse solvers. The joint spatio-temporal model leads to a cost function with an l(1)l(2)-norm regularizer whose minimization can be reduced to a convex second-order cone programming (SOCP) problem and efficiently solved using the interior-point method. The efficient computation of the SOCP problem allows us to implement permutation tests for estimating statistical significance of the inverse solution. Validation with simulated and human MEG data shows that the proposed solver yields source time course estimates qualitatively similar to those obtained through dipole fitting, but without the need to specify the number of dipole sources in advance. Furthermore, the l(1)l(2)-norm solver achieves fewer false positives and a better representation of the source locations than the conventional l(2) minimum-norm estimates.

Modeling adaptation effects in fMRI analysis.

The standard general linear model (GLM) for rapid event-related fMRI design protocols typically ignores reduction in hemodynamic responses in successive stimuli in a train due to incomplete recovery from the preceding stimuli. To capture this adaptation effect, we incorporate a region-specific adaptation model into GLM. The model quantifies the rate of adaptation across brain regions, which is of interest in neuroscience. Empirical evaluation of the proposed model demonstrates its potential to improve detection sensitivity. In the fMRI experiments using visual and auditory stimuli, we observed that the adaptation effect is significantly stronger in the visual area than in the auditory area, suggesting that we must account for this effect to avoid bias in fMRI detection.

A distributed spatio-temporal EEG/MEG inverse solver.

We propose a novel l1l2-norm inverse solver for estimating the sources of EEG/MEG signals. Based on the standard l1-norm inverse solver, the proposed sparse distributed inverse solver integrates the l1-norm spatial model with a temporal model of the source signals in order to avoid unstable activation patterns and "spiky" reconstructed signals often produced by the original solvers. The joint spatio-temporal model leads to a cost function with an l1l2-norm regularizer whose minimization can be reduced to a convex second-order cone programming problem and efficiently solved using the interior-point method. Validation with simulated and real MEG data shows that the proposed solver yields source time course estimates qualitatively similar to those obtained through dipole fitting, but without the need to specify the number of dipole sources in advance. Furthermore, the l1l2-norm solver achieves fewer false positives and a better representation of the source locations than the conventional l2 minimum-norm estimates.

On the construction of invertible filter banks on the 2-sphere.

The theories of signal sampling, filter banks, wavelets, and "overcomplete wavelets" are well established for the Euclidean spaces and are widely used in the processing and analysis of images. While recent advances have extended some filtering methods to spherical images, many key challenges remain. In this paper, we develop theoretical conditions for the invertibility of filter banks under continuous spherical convolution. Furthermore, we present an analogue of the Papoulis generalized sampling theorem on the 2-Sphere. We use the theoretical results to establish a general framework for the design of invertible filter banks on the sphere and demonstrate the approach with examples of self-invertible spherical wavelets and steerable pyramids. We conclude by examining the use of a self-invertible spherical steerable pyramid in a denoising experiment and discussing the computational complexity of the filtering framework.

Sources of variability in MEG.

This paper investigates and characterizes sources of variability in MEG signals in multi-site, multi-subject studies. Understanding these sources will help to develop efficient strategies for comparing and pooling data across repetitions of an experiment, across subjects, and across sites. In this work, we investigated somatosensory MEG data collected at three different sites and applied variance component analysis and nonparametric KL divergence analysis in order to characterize the sources of variability. Our analysis showed that inter-subject differences are the biggest factor in the signal variability. We demonstrated that the timing of the deflections is very consistent in the early somatosensory response, which justifies a direct comparison of deflection peak times acquired from different visits, subjects, and systems. Compared with deflection peak times, deflection magnitudes have larger variation across sites; modeling of this variability is necessary for data pooling.

From spatial regularization to anatomical priors in fMRI analysis.

In this paper, we study Markov Random Fields as spatial smoothing priors in fMRI detection. Relatively high noise in fMRI images presents a serious challenge for the detection algorithms, creating a need for spatial regularization of the signal. Gaussian smoothing, traditionally employed to boost the signal-to-noise ratio, often removes small activation regions. Recently, the use of Markov priors has been suggested as an alternative regularization approach. In this work, we investigate fast approximate inference algorithms for using MRFs in fMRI detection, propose a novel way to incorporate anatomical information into the detection framework, validate the methods through ROC analysis on simulated data and demonstrate their application in a real fMRI study.