Browsing by Subject "machine learning"

Now showing 1 - 20 of 22

Open Access
Algorithmic handwriting analysis of Judah's military correspondence sheds light on composition of biblical texts.
(Proceedings of the National Academy of Sciences of the United States of America, 2016-04) Faigenbaum-Golovin, Shira; Shaus, Arie; Sober, Barak; Levin, David; Na'aman, Nadav; Sass, Benjamin; Turkel, Eli; Piasetzky, Eli; Finkelstein, Israel
The relationship between the expansion of literacy in Judah and composition of biblical texts has attracted scholarly attention for over a century. Information on this issue can be deduced from Hebrew inscriptions from the final phase of the first Temple period. We report our investigation of 16 inscriptions from the Judahite desert fortress of Arad, dated ca 600 BCE-the eve of Nebuchadnezzar's destruction of Jerusalem. The inquiry is based on new methods for image processing and document analysis, as well as machine learning algorithms. These techniques enable identification of the minimal number of authors in a given group of inscriptions. Our algorithmic analysis, complemented by the textual information, reveals a minimum of six authors within the examined inscriptions. The results indicate that in this remote fort literacy had spread throughout the military hierarchy, down to the quartermaster and probably even below that rank. This implies that an educational infrastructure that could support the composition of literary texts in Judah already existed before the destruction of the first Temple. A similar level of literacy in this area is attested again only 400 y later, ca 200 BCE.
Open Access
An Ensemble Approach to Knowledge-Based Intensity-Modulated Radiation Therapy Planning.
(Frontiers in oncology, 2018-01) Zhang, Jiahan; Wu, Q Jackie; Xie, Tianyi; Sheng, Yang; Yin, Fang-Fang; Ge, Yaorong
Knowledge-based planning (KBP) utilizes experienced planners' knowledge embedded in prior plans to estimate optimal achievable dose volume histogram (DVH) of new cases. In the regression-based KBP framework, previously planned patients' anatomical features and DVHs are extracted, and prior knowledge is summarized as the regression coefficients that transform features to organ-at-risk DVH predictions. In our study, we find that in different settings, different regression methods work better. To improve the robustness of KBP models, we propose an ensemble method that combines the strengths of various linear regression models, including stepwise, lasso, elastic net, and ridge regression. In the ensemble approach, we first obtain individual model prediction metadata using in-training-set leave-one-out cross validation. A constrained optimization is subsequently performed to decide individual model weights. The metadata is also used to filter out impactful training set outliers. We evaluate our method on a fresh set of retrospectively retrieved anonymized prostate intensity-modulated radiation therapy (IMRT) cases and head and neck IMRT cases. The proposed approach is more robust against small training set size, wrongly labeled cases, and dosimetric inferior plans, compared with other individual models. In summary, we believe the improved robustness makes the proposed method more suitable for clinical settings than individual models.
Open Access
Application of Machine Learning in Pulmonary Function Assessment Where Are We Now and Where Are We Going?
(Frontiers in physiology, 2021-01) Giri, Paresh C; Chowdhury, Anand M; Bedoya, Armando; Chen, Hengji; Lee, Hyun Suk; Lee, Patty; Henriquez, Craig; MacIntyre, Neil R; Huang, Yuh-Chin T
Analysis of pulmonary function tests (PFTs) is an area where machine learning (ML) may benefit clinicians, researchers, and the patients. PFT measures spirometry, lung volumes, and carbon monoxide diffusion capacity of the lung (DLCO). The results are usually interpreted by the clinicians using discrete numeric data according to published guidelines. PFT interpretations by clinicians, however, are known to have inter-rater variability and the inaccuracy can impact patient care. This variability may be caused by unfamiliarity of the guidelines, lack of training, inadequate understanding of lung physiology, or simply mental lapses. A rules-based automated interpretation system can recapitulate expert's pattern recognition capability and decrease errors. ML can also be used to analyze continuous data or the graphics, including the flow-volume loop, the DLCO and the nitrogen washout curves. These analyses can discover novel physiological biomarkers. In the era of wearables and telehealth, particularly with the COVID-19 pandemic restricting PFTs to be done in the clinical laboratories, ML can also be used to combine mobile spirometry results with an individual's clinical profile to deliver precision medicine. There are, however, hurdles in the development and commercialization of the ML-assisted PFT interpretation programs, including the need for high quality representative data, the existence of different formats for data acquisition and sharing in PFT software by different vendors, and the need for collaboration amongst clinicians, biomedical engineers, and information technologists. Hurdles notwithstanding, the new developments would represent significant advances that could be the future of PFT, the oldest test still in use in clinical medicine.
Open Access
Applying machine learning to investigate long-term insect-plant interactions preserved on digitized herbarium specimens.
(Applications in plant sciences, 2020-06) Meineke, Emily K; Tomasi, Carlo; Yuan, Song; Pryer, Kathleen M
Premise:Despite the economic significance of insect damage to plants (i.e., herbivory), long-term data documenting changes in herbivory are limited. Millions of pressed plant specimens are now available online and can be used to collect big data on plant-insect interactions during the Anthropocene. Methods:We initiated development of machine learning methods to automate extraction of herbivory data from herbarium specimens by training an insect damage detector and a damage type classifier on two distantly related plant species (Quercus bicolor and Onoclea sensibilis). We experimented with (1) classifying six types of herbivory and two control categories of undamaged leaf, and (2) detecting two of the damage categories for which several hundred annotations were available. Results:Damage detection results were mixed, with a mean average precision of 45% in the simultaneous detection and classification of two types of damage. However, damage classification on hand-drawn boxes identified the correct type of herbivory 81.5% of the time in eight categories. The damage classifier was accurate for categories with 100 or more test samples. Discussion:These tools are a promising first step for the automation of herbivory data collection. We describe ongoing efforts to increase the accuracy of these models, allowing researchers to extract similar data and apply them to biological hypotheses.
Open Access
Automatic Planning of Whole Breast Radiation Therapy Using Machine Learning Models.
(Frontiers in Oncology, 2019-01) Sheng, Yang; Li, Taoran; Yoo, Sua; Yin, Fang-Fang; Blitzblau, Rachel; Horton, Janet K; Ge, Yaorong; Wu, Q Jackie
Purpose: To develop an automatic treatment planning system for whole breast radiation therapy (WBRT) based on two intensity-modulated tangential fields, enabling near-real-time planning. Methods and Materials: A total of 40 WBRT plans from a single institution were included in this study under IRB approval. Twenty WBRT plans, 10 with single energy (SE, 6MV) and 10 with mixed energy (ME, 6/15MV), were randomly selected as training dataset to develop the methodology for automatic planning. The rest 10 SE cases and 10 ME cases served as validation. The auto-planning process consists of three steps. First, an energy prediction model was developed to automate energy selection. This model establishes an anatomy-energy relationship based on principle component analysis (PCA) of the gray level histograms from training cases' digitally reconstructed radiographs (DRRs). Second, a random forest (RF) model generates an initial fluence map using the selected energies. Third, the balance of overall dose contribution throughout the breast tissue is realized by automatically selecting anchor points and applying centrality correction. The proposed method was tested on the validation dataset. Non-parametric equivalence test was performed for plan quality metrics using one-sided Wilcoxon Signed-Rank test. Results: For validation, the auto-planning system suggested same energy choices as clinical-plans in 19 out of 20 cases. The mean (standard deviation, SD) of percent target volume covered by 100% prescription dose was 82.5% (4.2%) for auto-plans, and 79.3% (4.8%) for clinical-plans (p > 0.999). Mean (SD) volume receiving 105% Rx were 95.2 cc (90.7 cc) for auto-plans and 83.9 cc (87.2 cc) for clinical-plans (p = 0.108). Optimization time for auto-plan was <20 s while clinical manual planning takes between 30 min and 4 h. Conclusions: We developed an automatic treatment planning system that generates WBRT plans with optimal energy selection, clinically comparable plan quality, and significant reduction in planning time, allowing for near-real-time planning.
Open Access
Current State of and Future Opportunities for Prediction in Microbiome Research: Report from the Mid-Atlantic Microbiome Meet-up in Baltimore on 9 January 2019.
(mSystems, 2019-10) Sakowski, Eric; Uritskiy, Gherman; Cooper, Rachel; Gomes, Maya; McLaren, Michael R; Meisel, Jacquelyn S; Mickol, Rebecca L; Mintz, C David; Mongodin, Emmanuel F; Pop, Mihai; Rahman, Mohammad Arifur; Sanchez, Alvaro; Timp, Winston; Vela, Jeseth Delgado; Wolz, Carly Muletz; Zackular, Joseph P; Chopyk, Jessica; Commichaux, Seth; Davis, Meghan; Dluzen, Douglas; Ganesan, Sukirth M; Haruna, Muyideen; Nasko, Dan; Regan, Mary J; Sarria, Saul; Shah, Nidhi; Stacy, Brook; Taylor, Dylan; DiRuggiero, Jocelyne; Preheim, Sarah P
Accurate predictions across multiple fields of microbiome research have far-reaching benefits to society, but there are few widely accepted quantitative tools to make accurate predictions about microbial communities and their functions. More discussion is needed about the current state of microbiome analysis and the tools required to overcome the hurdles preventing development and implementation of predictive analyses. We summarize the ideas generated by participants of the Mid-Atlantic Microbiome Meet-up in January 2019. While it was clear from the presentations that most fields have advanced beyond simple associative and descriptive analyses, most fields lack essential elements needed for the development and application of accurate microbiome predictions. Participants stressed the need for standardization, reproducibility, and accessibility of quantitative tools as key to advancing predictions in microbiome analysis. We highlight hurdles that participants identified and propose directions for future efforts that will advance the use of prediction in microbiome research.
Open Access
Development and Evaluation of a Small Airway Disease Index Derived From Modeling the Late-Expiratory Flattening of the Flow-Volume Loop.
(Frontiers in physiology, 2022-01) Chen, Hengji; Joshi, Sangeeta; Oberle, Amber J; Wong, An-Kwok; Shaz, David; Thapamagar, Suman; Tan, Laren; Anholm, James D; Giri, Paresh C; Henriquez, Craig; Huang, Yuh-Chin T
Excessive decrease in the flow of the late expiratory portion of a flow volume loop (FVL) or "flattening", reflects small airway dysfunction. The assessment of the flattening is currently determined by visual inspection by the pulmonary function test (PFT) interpreters and is highly variable. In this study, we developed an objective measure to quantify the flattening. We downloaded 172 PFT reports in PDF format from the electronic medical records and digitized and extracted the expiratory portion of the FVL. We located point A (the point of the peak expiratory flow), point B (the point corresponding to 75% of the expiratory vital capacity), and point C (the end of the expiratory portion of the FVL intersecting with the x-axis). We did a linear fitting to the A-B segment and the B-C segment. We calculated: 1) the AB-BC angle (∠ABC), 2) BC-x-axis angle (∠BCX), and 3) the log ratio of the BC slope over the vertical distance between point A and x-axis [log (BC/A-x)]. We asked an expert pulmonologist to assess the FVLs and separated the 172 PFTs into the flattening and the non-flattening groups. We defined the cutoff value as the mean minus one standard deviation using data from the non-flattening group. ∠ABC had the best concordance rate of 80.2% with a cutoff value of 149.7°. We then asked eight pulmonologists to evaluate the flattening with and without ∠ABC in another 168 PFTs. The Fleiss' kappa was 0.320 (lower and upper confidence intervals [CIs]: 0.293 and 0.348 respectively) without ∠ABC and increased to 0.522 (lower and upper CIs: 0.494 and 0.550) with ∠ABC. There were 147 CT scans performed within 6 months of the 172 PFTs. Twenty-six of 55 PFTs (47.3%) with ∠ABC <149.7° had CT scans showing small airway disease patterns while 44 of 92 PFTs (47.8%) with ∠ABC ≥149.7° had no CT evidence of small airway disease. We concluded that ∠ABC improved the inter-rater agreement on the presence of the late expiratory flattening in FVL. It could be a useful addition to the assessment of small airway disease in the PFT interpretation algorithm and reporting.
Open Access
Evaluation of ML-Based Clinical Decision Support Tool to Replace an Existing Tool in an Academic Health System: Lessons Learned.
(Journal of personalized medicine, 2020-08-27) Woo, Myung; Alhanti, Brooke; Lusk, Sam; Dunston, Felicia; Blackwelder, Stephen; Lytle, Kay S; Goldstein, Benjamin A; Bedoya, Armando
There is increasing application of machine learning tools to problems in healthcare, with an ultimate goal to improve patient safety and health outcomes. When applied appropriately, machine learning tools can augment clinical care provided to patients. However, even if a model has impressive performance characteristics, prospectively evaluating and effectively implementing models into clinical care remains difficult. The primary objective of this paper is to recount our experiences and challenges in comparing a novel machine learning-based clinical decision support tool to legacy, non-machine learning tools addressing potential safety events in the hospitals and to summarize the obstacles which prevented evaluation of clinical efficacy of tools prior to widespread institutional use. We collected and compared safety events data, specifically patient falls and pressure injuries, between the standard of care approach and machine learning (ML)-based clinical decision support (CDS). Our assessment was limited to performance of the model rather than the workflow due to challenges in directly comparing both approaches. We did note a modest improvement in falls with ML-based CDS; however, it was not possible to determine that overall improvement was due to model characteristics.
Open Access
Global Convergence of Localized Policy Iteration in Networked Multi-Agent Reinforcement Learning
(Proceedings of the ACM on Measurement and Analysis of Computing Systems, 2023-02-28) Zhang, Y; Qu, G; Xu, P; Lin, Y; Chen, Z; Wierman, A
We study a multi-agent reinforcement learning (MARL) problem where the agents interact over a given network. The goal of the agents is to cooperatively maximize the average of their entropy-regularized long-term rewards. To overcome the curse of dimensionality and to reduce communication, we propose a Localized Policy Iteration (LPI) algorithm that provably learns a near-globally-optimal policy using only local information. In particular, we show that, despite restricting each agent's attention to only its κ-hop neighborhood, the agents are able to learn a policy with an optimality gap that decays polynomially in κ. In addition, we show the finite-sample convergence of LPI to the global optimal policy, which explicitly captures the trade-off between optimality and computational complexity in choosing κ. Numerical simulations demonstrate the effectiveness of LPI.
Open Access
Machine Learning and Precision Medicine in Emergency Medicine: The Basics.
(Cureus, 2021-09) Lee, Sangil; Lam, Samuel H; Hernandes Rocha, Thiago Augusto; Fleischman, Ross J; Staton, Catherine A; Taylor, Richard; Limkakeng, Alexander T
As machine learning (ML) and precision medicine become more readily available and used in practice, emergency physicians must understand the potential advantages and limitations of the technology. This narrative review focuses on the key components of machine learning, artificial intelligence, and precision medicine in emergency medicine (EM). Based on the content expertise, we identified articles from EM literature. The authors provided a narrative summary of each piece of literature. Next, the authors provided an introduction of the concepts of ML, artificial intelligence as an extension of ML, and precision medicine. This was followed by concrete examples of their applications in practice and research. Subsequently, we shared our thoughts on how to consume the existing research in these subjects and conduct high-quality research for academic emergency medicine. We foresee that the EM community will continue to adapt machine learning, artificial intelligence, and precision medicine in research and practice. We described several key components using our expertise.
Open Access
Machine Learning Consensus Clustering Approach for Patients with Lactic Acidosis in Intensive Care Units.
(Journal of personalized medicine, 2021-11) Pattharanitima, Pattharawin; Thongprayoon, Charat; Petnak, Tananchai; Srivali, Narat; Gembillo, Guido; Kaewput, Wisit; Chesdachai, Supavit; Vallabhajosyula, Saraschandra; O'Corragain, Oisin A; Mao, Michael A; Garovic, Vesna D; Qureshi, Fawad; Dillon, John J; Cheungpasitporn, Wisit
Lactic acidosis is a heterogeneous condition with multiple underlying causes and associated outcomes. The use of multi-dimensional patient data to subtype lactic acidosis can personalize patient care. Machine learning consensus clustering may identify lactic acidosis subgroups with unique clinical profiles and outcomes. We used the Medical Information Mart for Intensive Care III database to abstract electronic medical record data from patients admitted to intensive care units (ICU) in a tertiary care hospital in the United States. We included patients who developed lactic acidosis (defined as serum lactate ≥ 4 mmol/L) within 48 h of ICU admission. We performed consensus clustering analysis based on patient characteristics, comorbidities, vital signs, organ supports, and laboratory data to identify clinically distinct lactic acidosis subgroups. We calculated standardized mean differences to show key subgroup features. We compared outcomes among subgroups. We identified 1919 patients with lactic acidosis. The algorithm revealed three best unique lactic acidosis subgroups based on patient variables. Cluster 1 (n = 554) was characterized by old age, elective admission to cardiac surgery ICU, vasopressor use, mechanical ventilation use, and higher pH and serum bicarbonate. Cluster 2 (n = 815) was characterized by young age, admission to trauma/surgical ICU with higher blood pressure, lower comorbidity burden, lower severity index, and less vasopressor use. Cluster 3 (n = 550) was characterized by admission to medical ICU, history of liver disease and coagulopathy, acute kidney injury, lower blood pressure, higher comorbidity burden, higher severity index, higher serum lactate, and lower pH and serum bicarbonate. Cluster 3 had the worst outcomes, while cluster 1 had the most favorable outcomes in terms of persistent lactic acidosis and mortality. Consensus clustering analysis synthesized the pattern of clinical and laboratory data to reveal clinically distinct lactic acidosis subgroups with different outcomes.
Open Access
Machine learning for early detection of sepsis: an internal and temporal validation study.
(JAMIA open, 2020-07) Bedoya, Armando D; Futoma, Joseph; Clement, Meredith E; Corey, Kristin; Brajer, Nathan; Lin, Anthony; Simons, Morgan G; Gao, Michael; Nichols, Marshall; Balu, Suresh; Heller, Katherine; Sendak, Mark; O'Brien, Cara
Objective
Determine if deep learning detects sepsis earlier and more accurately than other models. To evaluate model performance using implementation-oriented metrics that simulate clinical practice.
Materials and methods
We trained internally and temporally validated a deep learning model (multi-output Gaussian process and recurrent neural network [MGP-RNN]) to detect sepsis using encounters from adult hospitalized patients at a large tertiary academic center. Sepsis was defined as the presence of 2 or more systemic inflammatory response syndrome (SIRS) criteria, a blood culture order, and at least one element of end-organ failure. The training dataset included demographics, comorbidities, vital signs, medication administrations, and labs from October 1, 2014 to December 1, 2015, while the temporal validation dataset was from March 1, 2018 to August 31, 2018. Comparisons were made to 3 machine learning methods, random forest (RF), Cox regression (CR), and penalized logistic regression (PLR), and 3 clinical scores used to detect sepsis, SIRS, quick Sequential Organ Failure Assessment (qSOFA), and National Early Warning Score (NEWS). Traditional discrimination statistics such as the C-statistic as well as metrics aligned with operational implementation were assessed.
Results
The training set and internal validation included 42 979 encounters, while the temporal validation set included 39 786 encounters. The C-statistic for predicting sepsis within 4 h of onset was 0.88 for the MGP-RNN compared to 0.836 for RF, 0.849 for CR, 0.822 for PLR, 0.756 for SIRS, 0.619 for NEWS, and 0.481 for qSOFA. MGP-RNN detected sepsis a median of 5 h in advance. Temporal validation assessment continued to show the MGP-RNN outperform all 7 clinical risk score and machine learning comparisons.
Conclusions
We developed and validated a novel deep learning model to detect sepsis. Using our data elements and feature set, our modeling approach outperformed other machine learning methods and clinical scores.
Open Access
Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis.
(Journal of clinical medicine, 2021-10) Pattharanitima, Pattharawin; Thongprayoon, Charat; Kaewput, Wisit; Qureshi, Fawad; Qureshi, Fahad; Petnak, Tananchai; Srivali, Narat; Gembillo, Guido; O'Corragain, Oisin A; Chesdachai, Supavit; Vallabhajosyula, Saraschandra; Guru, Pramod K; Mao, Michael A; Garovic, Vesna D; Dillon, John J; Cheungpasitporn, Wisit
Lactic acidosis is the most common cause of anion gap metabolic acidosis in the intensive care unit (ICU), associated with poor outcomes including mortality. We sought to compare machine learning (ML) approaches versus logistic regression analysis for prediction of mortality in lactic acidosis patients admitted to the ICU. We used the Medical Information Mart for Intensive Care (MIMIC-III) database to identify ICU adult patients with lactic acidosis (serum lactate ≥4 mmol/L). The outcome of interest was hospital mortality. We developed prediction models using four ML approaches consisting of random forest (RF), decision tree (DT), extreme gradient boosting (XGBoost), artificial neural network (ANN), and statistical modeling with forward stepwise logistic regression using the testing dataset. We then assessed model performance using area under the receiver operating characteristic curve (AUROC), accuracy, precision, error rate, Matthews correlation coefficient (MCC), F1 score, and assessed model calibration using the Brier score, in the independent testing dataset. Of 1919 lactic acidosis ICU patients, 1535 and 384 were included in the training and testing dataset, respectively. Hospital mortality was 30%. RF had the highest AUROC at 0.83, followed by logistic regression 0.81, XGBoost 0.81, ANN 0.79, and DT 0.71. In addition, RF also had the highest accuracy (0.79), MCC (0.45), F1 score (0.56), and lowest error rate (21.4%). The RF model was the most well-calibrated. The Brier score for RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.15, 0.19, 0.18, 0.19, and 0.16, respectively. The RF model outperformed multivariable logistic regression model, SOFA score (AUROC 0.74), SAP II score (AUROC 0.77), and Charlson score (AUROC 0.69). The ML prediction model using RF algorithm provided the highest predictive performance for hospital mortality among ICU patient with lactic acidosis.
Open Access
Overcoming barriers to the adoption and implementation of predictive modeling and machine learning in clinical care: what can we learn from US academic medical centers?
(JAMIA open, 2020-07) Watson, Joshua; Hutyra, Carolyn A; Clancy, Shayna M; Chandiramani, Anisha; Bedoya, Armando; Ilangovan, Kumar; Nderitu, Nancy; Poon, Eric G
There is little known about how academic medical centers (AMCs) in the US develop, implement, and maintain predictive modeling and machine learning (PM and ML) models. We conducted semi-structured interviews with leaders from AMCs to assess their use of PM and ML in clinical care, understand associated challenges, and determine recommended best practices. Each transcribed interview was iteratively coded and reconciled by a minimum of 2 investigators to identify key barriers to and facilitators of PM and ML adoption and implementation in clinical care. Interviews were conducted with 33 individuals from 19 AMCs nationally. AMCs varied greatly in the use of PM and ML within clinical care, from some just beginning to explore their utility to others with multiple models integrated into clinical care. Informants identified 5 key barriers to the adoption and implementation of PM and ML in clinical care: (1) culture and personnel, (2) clinical utility of the PM and ML tool, (3) financing, (4) technology, and (5) data. Recommendation to the informatics community to overcome these barriers included: (1) development of robust evaluation methodologies, (2) partnership with vendors, and (3) development and dissemination of best practices. For institutions developing clinical PM and ML applications, they are advised to: (1) develop appropriate governance, (2) strengthen data access, integrity, and provenance, and (3) adhere to the 5 rights of clinical decision support. This article highlights key challenges of implementing PM and ML in clinical care at AMCs and suggests best practices for development, implementation, and maintenance at these institutions.
Open Access
Predicting the risk of rupture for vertebral aneurysm based on geometric features of blood vessels.
(Royal Society open science, 2021-08-11) Li, Shixuan; Pan, Ruiqi; Gupta, Arvind; Xu, Shixin; Fang, Yibin; Huang, Huaxiong
A significant proportion of the adult population worldwide suffers from cerebral aneurysms. If left untreated, aneurysms may rupture and lead to fatal massive internal bleeding. On the other hand, treatment of aneurysms also involve significant risks. It is desirable, therefore, to have an objective tool that can be used to predict the risk of rupture and assist in surgical decision for operating on the aneurysms. Currently, such decisions are made mostly based on medical expertise of the healthcare team. In this paper, we investigate the possibility of using machine learning algorithms to predict rupture risk of vertebral artery fusiform aneurysms based on geometric features of the blood vessels surrounding but excluding the aneurysm. For each of the aneurysm images (12 ruptured and 25 unruptured), the vessel is segmented into distal and proximal parts by cross-sectional area and 382 non-aneurysm-related geometric features extracted. The decision tree model using two of the features (standard deviation of eccentricity of proximal vessel, and diameter at the distal endpoint) achieved 83.8% classification accuracy. Additionally, with support vector machine and logistic regression, we also achieved 83.8% accuracy with another set of two features (ratio of mean curvature between distal and proximal parts, and diameter at the distal endpoint). Combining the aforementioned three features with integration of curvature of proximal vessel and also ratio of mean cross-sectional area between distal and proximal parts, these models achieve an impressive 94.6% accuracy. These results strongly suggest the usefulness of geometric features in predicting the risk of rupture.
Open Access
Seabird trophic position across three ocean regions tracks ecosystem differences
(Frontiers in Marine Science, 2018-09-07) Gagné, TO; Hyrenbach, KD; Hagemann, ME; Bass, OL; Pimm, SL; MacDonald, M; Peck, B; Van Houtan, KS
We analyze recently collected feather tissues from two species of seabirds, the sooty tern (Onychoprion fuscatus) and brown noddy (Anous stolidus), in three ocean regions (North Atlantic, North Pacific, and South Pacific) with different human impacts. The species are similar morphologically and in the trophic levels from which they feed within each location. In contrast, we detect reliable differences in trophic position amongst the regions. Trophic position appears to decline as the intensity of commercial fishing increases, and is at its lowest in the Caribbean. The spatial gradient in trophic position we document in these regions exceeds those detected over specimens from the last 130 years in the Hawaiian Islands. Modeling suggests that climate velocity and human impacts on fish populations strongly align with these differences.
Open Access
Testing for Unobserved Heterogeneity via K-Means Clustering
(2019-07-15) Patton, AJ; Weller, BM
Open Access
Unbiased kidney-centric molecular categorization of chronic kidney disease as a step towards precision medicine.
(Kidney international, 2024-01) Reznichenko, Anna; Nair, Viji; Eddy, Sean; Fermin, Damian; Tomilo, Mark; Slidel, Timothy; Ju, Wenjun; Henry, Ian; Badal, Shawn S; Wesley, Johnna D; Liles, John T; Moosmang, Sven; Williams, Julie M; Quinn, Carol Moreno; Bitzer, Markus; Hodgin, Jeffrey B; Barisoni, Laura; Karihaloo, Anil; Breyer, Matthew D; Duffin, Kevin L; Patel, Uptal D; Magnone, Maria Chiara; Bhat, Ratan; Kretzler, Matthias
Current classification of chronic kidney disease (CKD) into stages using indirect systemic measures (estimated glomerular filtration rate (eGFR) and albuminuria) is agnostic to the heterogeneity of underlying molecular processes in the kidney thereby limiting precision medicine approaches. To generate a novel CKD categorization that directly reflects within kidney disease drivers we analyzed publicly available transcriptomic data from kidney biopsy tissue. A Self-Organizing Maps unsupervised artificial neural network machine-learning algorithm was used to stratify a total of 369 patients with CKD and 46 living kidney donors as healthy controls. Unbiased stratification of the discovery cohort resulted in identification of four novel molecular categories of disease termed CKD-Blue, CKD-Gold, CKD-Olive, CKD-Plum that were replicated in independent CKD and diabetic kidney disease datasets and can be further tested on any external data at kidneyclass.org. Each molecular category spanned across CKD stages and histopathological diagnoses and represented transcriptional activation of distinct biological pathways. Disease progression rates were highly significantly different between the molecular categories. CKD-Gold displayed rapid progression, with significant eGFR-adjusted Cox regression hazard ratio of 5.6 [1.01-31.3] for kidney failure and hazard ratio of 4.7 [1.3-16.5] for composite of kidney failure or a 40% or more eGFR decline. Urine proteomics revealed distinct patterns between the molecular categories, and a 25-protein signature was identified to distinguish CKD-Gold from other molecular categories. Thus, patient stratification based on kidney tissue omics offers a gateway to non-invasive biomarker-driven categorization and the potential for future clinical implementation, as a key step towards precision medicine in CKD.
Open Access
Universal Digital High Resolution Melt for the detection of pulmonary mold infections.
(bioRxiv, 2023-11-09) Goshia, Tyler; Aralar, April; Wiederhold, Nathan; Jenks, Jeffrey D; Mehta, Sanjay R; Sinha, Mridu; Karmakar, Aprajita; Sharma, Ankit; Shrivastava, Rachit; Sun, Haoxiang; White, P Lewis; Hoenigl, Martin; Fraley, Stephanie I
BACKGROUND: Invasive mold infections (IMIs) such as aspergillosis, mucormycosis, fusariosis, and lomentosporiosis are associated with high morbidity and mortality, particularly in immunocompromised patients, with mortality rates as high as 40% to 80%. Outcomes could be substantially improved with early initiation of appropriate antifungal therapy, yet early diagnosis remains difficult to establish and often requires multidisciplinary teams evaluating clinical and radiological findings plus supportive mycological findings. Universal digital high resolution melting analysis (U-dHRM) may enable rapid and robust diagnosis of IMI. This technology aims to accomplish timely pathogen detection at the single genome level by conducting broad-based amplification of microbial barcoding genes in a digital polymerase chain reaction (dPCR) format, followed by high-resolution melting of the DNA amplicons in each digital reaction to generate organism-specific melt curve signatures that are identified by machine learning. METHODS: A universal fungal assay was developed for U-dHRM and used to generate a database of melt curve signatures for 19 clinically relevant fungal pathogens. A machine learning algorithm (ML) was trained to automatically classify these 19 fungal melt curves and detect novel melt curves. Performance was assessed on 73 clinical bronchoalveolar lavage (BAL) samples from patients suspected of IMI. Novel curves were identified by micropipetting U-dHRM reactions and Sanger sequencing amplicons. RESULTS: U-dHRM achieved an average of 97% fungal organism identification accuracy and a turn-around-time of 4hrs. Pathogenic molds (Aspergillus, Mucorales, Lomentospora and Fusarium) were detected by U-dHRM in 73% of BALF samples suspected of IMI. Mixtures of pathogenic molds were detected in 19%. U-dHRM demonstrated good sensitivity for IMI, as defined by current diagnostic criteria, when clinical findings were also considered. CONCLUSIONS: U-dHRM showed promising performance as a separate or combination diagnostic approach to standard mycological tests. The speed of U-dHRM and its ability to simultaneously identify and quantify clinically relevant mold pathogens in polymicrobial samples as well as detect emerging opportunistic pathogens may provide information that could aid in treatment decisions and improve patient outcomes.
Open Access
Using computer vision on herbarium specimen images to discriminate among closely related horsetails (Equisetum).
(Applications in plant sciences, 2020-06) Pryer, KM; Tomasi, C; Wang, X; Meineke, EK; Windham, MD
Premise:Equisetum is a distinctive vascular plant genus with 15 extant species worldwide. Species identification is complicated by morphological plasticity and frequent hybridization events, leading to a disproportionately high number of misidentified specimens. These may be correctly identified by applying appropriate computer vision tools. Methods:We hypothesize that aerial stem nodes can provide enough information to distinguish among Equisetum hyemale, E. laevigatum, and E . ×ferrissii, the latter being a hybrid between the other two. An object detector was trained to find nodes on a given image and to distinguish E. hyemale nodes from those of E. laevigatum. A classifier then took statistics from the detection results and classified the given image into one of the three taxa. Both detector and classifier were trained and tested on expert manually annotated images. Results:In our exploratory test set of 30 images, our detector/classifier combination identified all 10 E. laevigatum images correctly, as well as nine out of 10 E. hyemale images, and eight out of 10 E. ×ferrissii images, for a 90% classification accuracy. Discussion:Our results support the notion that computer vision may help with the identification of herbarium specimens once enough manual annotations become available.

Browsing by Subject "machine learning"

Results Per Page

Sort Options

Objective

Materials and methods

Results

Conclusions