RESUMO
Background: Early identification of impending in-hospital cardiac arrest (IHCA) improves clinical outcomes but remains elusive for practicing clinicians. Objective: We aimed to develop a multimodal machine learning algorithm based on ensemble techniques to predict the occurrence of IHCA. Methods: Our model was developed by the Multiparameter Intelligent Monitoring of Intensive Care (MIMIC)-IV database and validated in the Electronic Intensive Care Unit Collaborative Research Database (eICU-CRD). Baseline features consisting of patient demographics, presenting illness, and comorbidities were collected to train a random forest model. Next, vital signs were extracted to train a long short-term memory model. A support vector machine algorithm then stacked the results to form the final prediction model. Results: Of 23,909 patients in the MIMIC-IV database and 10,049 patients in the eICU-CRD database, 452 and 85 patients, respectively, had IHCA. At 13 hours in advance of an IHCA event, our algorithm had already demonstrated an area under the receiver operating characteristic curve of 0.85 (95% CI 0.815-0.885) in the MIMIC-IV database. External validation with the eICU-CRD and National Taiwan University Hospital databases also presented satisfactory results, showing area under the receiver operating characteristic curve values of 0.81 (95% CI 0.763-0.851) and 0.945 (95% CI 0.934-0.956), respectively. Conclusions: Using only vital signs and information available in the electronic medical record, our model demonstrates it is possible to detect a trajectory of clinical deterioration up to 13 hours in advance. This predictive tool, which has undergone external validation, could forewarn and help clinicians identify patients in need of assessment to improve their overall prognosis.
RESUMO
In low- and middle-income countries, the substantial costs associated with traditional data collection pose an obstacle to facilitating decision-making in the field of public health. Satellite imagery offers a potential solution, but the image extraction and analysis can be costly and requires specialized expertise. We introduce SatelliteBench, a scalable framework for satellite image extraction and vector embeddings generation. We also propose a novel multimodal fusion pipeline that utilizes a series of satellite imagery and metadata. The framework was evaluated generating a dataset with a collection of 12,636 images and embeddings accompanied by comprehensive metadata, from 81 municipalities in Colombia between 2016 and 2018. The dataset was then evaluated in 3 tasks: including dengue case prediction, poverty assessment, and access to education. The performance showcases the versatility and practicality of SatelliteBench, offering a reproducible, accessible and open tool to enhance decision-making in public health.
Assuntos
Dengue , Saúde Pública , Imagens de Satélites , Colômbia , Humanos , MetadadosRESUMO
BACKGROUND: This study investigates the potential of diffusion tensor imaging (DTI) in identifying penumbral volume (PV) compared to the standard gadolinium-required perfusion-diffusion mismatch (PDM), utilizing a stack-based ensemble machine learning (ML) approach with enhanced explainability. METHODS: Sixteen male rats were subjected to middle cerebral artery occlusion. The penumbra was identified using PDM at 30 and 90 min after occlusion. We used 11 DTI-derived metrics and 14 distance-based features to train five voxel-wise ML models. The model predictions were integrated using stack-based ensemble techniques. ML-estimated and PDM-defined PVs were compared to evaluate model performance through volume similarity assessment, the Pearson correlation analysis, and Bland-Altman analysis. Feature importance was determined for explainability. RESULTS: In the test rats, the ML-estimated median PV was 106.4 mL (interquartile range 44.6-157.3 mL), whereas the PDM-defined median PV was 102.0 mL (52.1-144.9 mL). These PVs had a volume similarity of 0.88 (0.79-0.96), a Pearson correlation coefficient of 0.93 (p < 0.001), and a Bland-Altman bias of 2.5 mL (2.4% of the mean PDM-defined PV), with 95% limits of agreement ranging from -44.9 to 49.9 mL. Among the features used for PV prediction, the mean diffusivity was the most important feature. CONCLUSIONS: Our study confirmed that PV can be estimated using DTI metrics with a stack-based ensemble ML approach, yielding results comparable to the volume defined by the standard PDM. The model explainability enhanced its clinical relevance. Human studies are warranted to validate our findings. RELEVANCE STATEMENT: The proposed DTI-based ML model can estimate PV without the need for contrast agent administration, offering a valuable option for patients with kidney dysfunction. It also can serve as an alternative if perfusion map interpretation fails in the clinical setting. KEY POINTS: ⢠Penumbral volume can be estimated by DTI combined with stack-based ensemble ML. ⢠Mean diffusivity was the most important feature used for predicting penumbral volume. ⢠The proposed approach can be beneficial for patients with kidney dysfunction.
Assuntos
Imagem de Tensor de Difusão , Aprendizado de Máquina , Animais , Masculino , Ratos , Imagem de Tensor de Difusão/métodos , Infarto da Artéria Cerebral Média/diagnóstico por imagem , Ratos Sprague-DawleyRESUMO
BACKGROUND: It has been shown that AI models can learn race on medical images, leading to algorithmic bias. Our aim in this study was to enhance the fairness of medical image models by eliminating bias related to race, age, and sex. We hypothesise models may be learning demographics via shortcut learning and combat this using image augmentation. METHODS: This study included 44,953 patients who identified as Asian, Black, or White (mean age, 60.68 years ±18.21; 23,499 women) for a total of 194,359 chest X-rays (CXRs) from MIMIC-CXR database. The included CheXpert images comprised 45,095 patients (mean age 63.10 years ±18.14; 20,437 women) for a total of 134,300 CXRs were used for external validation. We also collected 1195 3D brain magnetic resonance imaging (MRI) data from the ADNI database, which included 273 participants with an average age of 76.97 years ±14.22, and 142 females. DL models were trained on either non-augmented or augmented images and assessed using disparity metrics. The features learned by the models were analysed using task transfer experiments and model visualisation techniques. FINDINGS: In the detection of radiological findings, training a model using augmented CXR images was shown to reduce disparities in error rate among racial groups (-5.45%), age groups (-13.94%), and sex (-22.22%). For AD detection, the model trained with augmented MRI images was shown 53.11% and 31.01% reduction of disparities in error rate among age and sex groups, respectively. Image augmentation led to a reduction in the model's ability to identify demographic attributes and resulted in the model trained for clinical purposes incorporating fewer demographic features. INTERPRETATION: The model trained using the augmented images was less likely to be influenced by demographic information in detecting image labels. These results demonstrate that the proposed augmentation scheme could enhance the fairness of interpretations by DL models when dealing with data from patients with different demographic backgrounds. FUNDING: National Science and Technology Council (Taiwan), National Institutes of Health.
Assuntos
Benchmarking , Aprendizagem , Idoso , Feminino , Humanos , Pessoa de Meia-Idade , População Negra , Encéfalo , Demografia , Estados Unidos , Povo Asiático , População Branca , Masculino , Negro ou Afro-AmericanoRESUMO
This study applied machine learning for the early prediction of 30-day mortality at sepsis diagnosis time in critically ill patients. Retrospective study using data collected from the Medical Information Mart for Intensive Care IV database. The data of the patient cohort was divided on the basis of the year of hospitalization, into training (2008-2013), validation (2014-2016), and testing (2017-2019) datasets. 24,377 patients with the sepsis diagnosis time < 24 h after intensive care unit (ICU) admission were included. A gradient boosting tree-based algorithm (XGBoost) was used for training the machine learning model to predict 30-day mortality at sepsis diagnosis time in critically ill patients. Model performance was measured in both discrimination and calibration aspects. The model was interpreted using the SHapley Additive exPlanations (SHAP) module. The 30-day mortality rate of the testing dataset was 17.9%, and 39 features were selected for the machine learning model. Model performance on the testing dataset achieved an area under the receiver operating characteristic curve (AUROC) of 0.853 (95% CI 0.837-0.868) and an area under the precision-recall curves of 0.581 (95% CI 0.541-0.619). The calibration plot for the model revealed a slope of 1.03 (95% CI 0.94-1.12) and intercept of 0.14 (95% CI 0.04-0.25). The SHAP revealed the top three most significant features, namely age, increased red blood cell distribution width, and respiratory rate. Our study demonstrated the feasibility of using the interpretable machine learning model to predict mortality at sepsis diagnosis time.
Assuntos
Estado Terminal , Sepse , Humanos , Estudos Retrospectivos , Sepse/diagnóstico , Algoritmos , Aprendizado de MáquinaRESUMO
A rapidly aging world population is fueling a concomitant increase in Alzheimer's disease (AD) and related dementias (ADRD). Scientific inquiry, however, has largely focused on White populations in Australia, the European Union, and North America. As such, there is an incomplete understanding of AD in other populations. In this perspective, we describe research efforts and challenges of cohort studies from three regions of the world: Central America, East Africa, and East Asia. These cohorts are engaging with the Davos Alzheimer's Collaborative (DAC), a global partnership that brings together cohorts from around the world to advance understanding of AD. Each cohort is poised to leverage the widespread use of mobile devices to integrate digital phenotyping into current methodologies and mitigate the lack of representativeness in AD research of racial and ethnic minorities across the globe. In addition to methods that these three cohorts are already using, DAC has developed a digital phenotyping protocol that can collect ADRD-related data remotely via smartphone and/or in clinic via a tablet to generate a common data elements digital dataset that can be harmonized with additional clinical and molecular data being collected at each cohort site and when combined across cohorts and made accessible can provide a global data resource that is more racially/ethnically represented of the world population.
RESUMO
PURPOSE: Chronic obstructive pulmonary disease (COPD) is one of the most common chronic illnesses in the world. Unfortunately, COPD is often difficult to diagnose early when interventions can alter the disease course, and it is underdiagnosed or only diagnosed too late for effective treatment. Currently, spirometry is the gold standard for diagnosing COPD but it can be challenging to obtain, especially in resource-poor countries. Chest X-rays (CXRs), however, are readily available and may have the potential as a screening tool to identify patients with COPD who should undergo further testing or intervention. In this study, we used three CXR datasets alongside their respective electronic health records (EHR) to develop and externally validate our models. METHOD: To leverage the performance of convolutional neural network models, we proposed two fusion schemes: (1) model-level fusion, using Bootstrap aggregating to aggregate predictions from two models, (2) data-level fusion, using CXR image data from different institutions or multi-modal data, CXR image data, and EHR data for model training. Fairness analysis was then performed to evaluate the models across different demographic groups. RESULTS: Our results demonstrate that DL models can detect COPD using CXRs with an area under the curve of over 0.75, which could facilitate patient screening for COPD, especially in low-resource regions where CXRs are more accessible than spirometry. CONCLUSIONS: By using a ubiquitous test, future research could build on this work to detect COPD in patients early who would not otherwise have been diagnosed or treated, altering the course of this highly morbid disease.
RESUMO
PURPOSE: Despite the importance of radial endobronchial ultrasound (rEBUS) in transbronchial biopsy, researchers have yet to apply artificial intelligence to the analysis of rEBUS images. MATERIALS AND METHODS: This study developed a convolutional neural network (CNN) to differentiate between malignant and benign tumours in rEBUS images. This study retrospectively collected rEBUS images from medical centres in Taiwan, including 769 from National Taiwan University Hospital Hsin-Chu Branch, Hsinchu Hospital for model training (615 images) and internal validation (154 images) as well as 300 from National Taiwan University Hospital (NTUH-TPE) and 92 images were obtained from National Taiwan University Hospital Hsin-Chu Branch, Biomedical Park Hospital (NTUH-BIO) for external validation. Further assessments of the model were performed using image augmentation in the training phase and test-time augmentation (TTA). RESULTS: Using the internal validation dataset, the results were as follows: area under the curve (AUC) (0.88 (95% CI 0.83 to 0.92)), sensitivity (0.80 (95% CI 0.73 to 0.88)), specificity (0.75 (95% CI 0.66 to 0.83)). Using the NTUH-TPE external validation dataset, the results were as follows: AUC (0.76 (95% CI 0.71 to 0.80)), sensitivity (0.58 (95% CI 0.50 to 0.65)), specificity (0.92 (95% CI 0.88 to 0.97)). Using the NTUH-BIO external validation dataset, the results were as follows: AUC (0.72 (95% CI 0.64 to 0.82)), sensitivity (0.71 (95% CI 0.55 to 0.86)), specificity (0.76 (95% CI 0.64 to 0.87)). After fine-tuning, the AUC values for the external validation cohorts were as follows: NTUH-TPE (0.78) and NTUH-BIO (0.82). Our findings also demonstrated the feasibility of the model in differentiating between lung cancer subtypes, as indicated by the following AUC values: adenocarcinoma (0.70; 95% CI 0.64 to 0.76), squamous cell carcinoma (0.64; 95% CI 0.54 to 0.74) and small cell lung cancer (0.52; 95% CI 0.32 to 0.72). CONCLUSIONS: Our results demonstrate the feasibility of the proposed CNN-based algorithm in differentiating between malignant and benign lesions in rEBUS images.
Assuntos
Aprendizado Profundo , Neoplasias Pulmonares , Humanos , Inteligência Artificial , Estudos Retrospectivos , Redes Neurais de Computação , Neoplasias Pulmonares/diagnóstico por imagemRESUMO
Developing an efficient and environment-friendly route for waste valorization is extremely significant in accelerating the transition toward a circular economy. A novel waste-to-synthetic natural gas (SNG) conversion process comprising hybrid renewable energy systems is proposed for this purpose. This includes thermochemical waste conversion and power-to-gas technologies for simultaneous waste utilization and renewable energy storage applications. The energy and environmental performances of the proposed waste-to-SNG plant are assessed and optimized. Results indicated that the implementation of a thermal pretreatment unit prior to the plasma gasification (two-step) is beneficial to improve the yield of hydrogen in the syngas, thereby leading to less renewable energy requirement for green hydrogen production used in the methanation process. This also enhances SNG yield by a factor of 30% as compared to the case without thermal pretreatment (one-step). The overall energy efficiency (OE) of the proposed waste-to-SNG plant is in the range of 61.36-77.73%, while the energy return on investment (EROI) ranges between 2.66 and 6.11. Most environmental impacts are mainly contributed by the indirect carbon emissions as a consequence of the power requirement for thermal pretreatment, plasma gasifier, and auxiliary equipment. The value of specific electricity consumption for SNG production of the treated RDF exhibits 1.70-9.25 % less than that of raw RDF when the pretreatment temperature is less than 300 °C. The OE of the system declines by 4.52% when 50 wt% of biomass is mixed in the fuel, whereas an enhancement of 18.33% in EROI and a reduction of 16.19% in specific CO2 emissions are obtained.
Assuntos
Gás Natural , Eliminação de Resíduos , Eliminação de Resíduos/métodos , Energia Renovável , Temperatura , HidrogênioRESUMO
BACKGROUND: Timely differentiating between pulmonary tuberculosis (TB) and nontuberculous mycobacterial lung disease (NTM-LD), which are radiographically similar, is important because infectiousness and treatment differ. This study aimed to evaluate whether artificial intelligence could distinguish between TB or NTM-LD patients by chest X-rays (CXRs) from suspects of mycobacterial lung disease. METHODS: A total of 1500 CXRs, including 500 each from patients with pulmonary TB, NTM-LD, and patients with clinical suspicion but negative mycobacterial culture (Imitator) from two hospitals, were retrospectively collected and evaluated in this study. We developed a deep neural network (DNN) and evaluated model performance using the area under the receiver operating characteristic curves (AUC) in both internal and external test sets. Furthermore, we conducted a reader study and tested our model under three scenarios of different mycobacteria prevalence. RESULTS: Among the internal and external test sets, the AUCs of our DNN model were 0.83 ± 0.005 and 0.76 ± 0.006 for pulmonary TB, 0.86 ± 0.006 and 0.64 ± 0.017 for NTM-LD, and 0.77 ± 0.007 and 0.74 ± 0.005 for Imitator. The DNN model showed higher performance on the internal test set in classification accuracy (66.5 ± 2.5%) than senior (50.8 ± 3.0%, p < 0.001) and junior pulmonologists (47.5 ± 2.8%, p < 0.001). Among different prevalence scenarios, the DNN model has stable performance in terms of AUC to detect TB and mycobacterial lung disease. CONCLUSION: DNN model had satisfactory performance and a higher accuracy than pulmonologists on classifying patients with presumptive mycobacterial lung diseases. DNN model could be a complementary first-line screening tool.
RESUMO
OBJECTIVE: This study developed a diagnostic tool combining machine learning (ML) segmentation and radiomic texture analysis (RTA) for bone density screening using chest low-dose computed tomography (LDCT). METHODS: A total of 197 patients who underwent LDCT followed by dual-energy X-ray absorptiometry were analyzed. First, an autosegmentation model was trained using LDCT to delineate the thoracic vertebral body (VB). Second, a two-level classifier was developed using radiomic features extracted from VBs for the hierarchical pairwise classification of each patient's bone status. All the patients were initially classified as either normal or abnormal, and all patients with abnormal bone density were then subdivided into an osteopenia group and an osteoporosis group. The performance of the classifier was evaluated through fivefold cross-validation. RESULTS: The model for automated VB segmentation achieved a Sorenson-Dice coefficient of 0.87 ± 0.01. Furthermore, the area under the receiver operating characteristic curve scores for the two-level classifier were 0.96 ± 0.01 for detecting abnormal bone density (accuracy = 0.91 ± 0.02; sensitivity = 0.93 ± 0.03; specificity = 0.89 ± 0.03) and 0.98 ± 0.01 for distinguishing osteoporosis (accuracy = 0.94 ± 0.02; sensitivity = 0.95 ± 0.03; specificity = 0.93 ± 0.03). The testing prediction accuracy levels for the first- and second-level classifiers were 0.92 ± 0.04 and 0.94 ± 0.05, respectively. The overall testing prediction accuracy of our method was 0.90 ± 0.05. CONCLUSION: The combination of ML segmentation and RTA for automated bone density prediction based on LDCT scans is a feasible approach that could be valuable for osteoporosis screening during lung cancer screening. KEY POINTS: ⢠This study developed an automatic diagnostic tool combining machine learning-based segmentation and radiomic texture analysis for bone density screening using chest low-dose computed tomography. ⢠The developed method enables opportunistic screening without quantitative computed tomography or a dedicated phantom. ⢠The developed method could be integrated into the current clinical workflow and used as an adjunct for opportunistic screening or for patients who are ineligible for screening with dual-energy X-ray absorptiometry.
Assuntos
Neoplasias Pulmonares , Osteoporose , Humanos , Detecção Precoce de Câncer , Osteoporose/diagnóstico por imagem , Tomografia Computadorizada por Raios X/métodos , Densidade Óssea , Estudos RetrospectivosRESUMO
BACKGROUND: Multiple organ dysfunction syndrome (MODS) is associated with a high risk of mortality among older patients. Current severity scores are limited in their ability to assist clinicians with triage and management decisions. We aim to develop mortality prediction models for older patients with MODS admitted to the ICU. METHODS: The study analyzed older patients from 197 hospitals in the United States and 1 hospital in the Netherlands. The cohort was divided into the young-old (65-80 years) and old-old (≥80 years), which were separately used to develop and evaluate models including internal, external, and temporal validation. Demographic characteristics, comorbidities, vital signs, laboratory measurements, and treatments were used as predictors. We used the XGBoost algorithm to train models, and the SHapley Additive exPlanations (SHAP) method to interpret predictions. RESULTS: Thirty-four thousand four hundred and ninety-seven young-old (11.3% mortality) and 21 330 old-old (15.7% mortality) patients were analyzed. Discrimination AUROC of internal validation models in 9 046 U.S. patients was as follows: 0.87 and 0.82, respectively; discrimination of external validation models in 1 905 EUR patients was as follows: 0.86 and 0.85, respectively; and discrimination of temporal validation models in 8 690 U.S. patients: 0.85 and 0.78, respectively. These models outperformed standard clinical scores like Sequential Organ Failure Assessment and Acute Physiology Score III. The Glasgow Coma Scale, Charlson Comorbidity Index, and Code Status emerged as top predictors of mortality. CONCLUSIONS: Our models integrate data spanning physiologic and geriatric-relevant variables that outperform existing scores used in older adults with MODS, which represents a proof of concept of how machine learning can streamline data analysis for busy ICU clinicians to potentially optimize prognostication and decision making.
Assuntos
Hospitais , Insuficiência de Múltiplos Órgãos , Humanos , Idoso , Estudos Retrospectivos , Insuficiência de Múltiplos Órgãos/diagnóstico , Mortalidade Hospitalar , Aprendizado de MáquinaRESUMO
The unprecedented global crisis brought about by the COVID-19 pandemic has sparked numerous efforts to create predictive models for the detection and prognostication of SARS-CoV-2 infections with the goal of helping health systems allocate resources. Machine learning models, in particular, hold promise for their ability to leverage patient clinical information and medical images for prediction. However, most of the published COVID-19 prediction models thus far have little clinical utility due to methodological flaws and lack of appropriate validation. In this paper, we describe our methodology to develop and validate multi-modal models for COVID-19 mortality prediction using multi-center patient data. The models for COVID-19 mortality prediction were developed using retrospective data from Madrid, Spain (N = 2547) and were externally validated in patient cohorts from a community hospital in New Jersey, USA (N = 242) and an academic center in Seoul, Republic of Korea (N = 336). The models we developed performed differently across various clinical settings, underscoring the need for a guided strategy when employing machine learning for clinical decision-making. We demonstrated that using features from both the structured electronic health records and chest X-ray imaging data resulted in better 30-day mortality prediction performance across all three datasets (areas under the receiver operating characteristic curves: 0.85 (95% confidence interval: 0.83-0.87), 0.76 (0.70-0.82), and 0.95 (0.92-0.98)). We discuss the rationale for the decisions made at every step in developing the models and have made our code available to the research community. We employed the best machine learning practices for clinical model development. Our goal is to create a toolkit that would assist investigators and organizations in building multi-modal models for prediction, classification, and/or optimization.
Assuntos
COVID-19 , Humanos , Estudos Retrospectivos , Pandemias , SARS-CoV-2 , Aprendizado de MáquinaRESUMO
Objective. Functional magnetic resonance imaging (fMRI) requires thresholds by which to identify brain regions with significant activation, particularly for experiments involving real-life paradigms. One conventional non-parametric approach to generating surrogate data involves decomposition of the original fMRI time series using the Fourier transform, after which the phase is randomized without altering the magnitude of individual frequency components. However, it has been reported that spontaneous brain signals could be non-stationary, which, if true, could lead to false-positive results.Approach. This paper introduces a randomization procedure based on the Hilbert-Huang transform by which to account for non-stationarity in fMRI time series derived from two fMRI datasets (stationary or non-stationary). The significance of individual voxels was determined by comparing the distribution of empirical data versus a surrogate distribution.Main results. In a comparison with conventional phase-randomization and wavelet-based permutation methods, the proposed method proved highly effective in generating activation maps indicating essential brain regions, while filtering out noise in the white matter.Significance. This work demonstrated the importance of considering the non-stationary nature of fMRI time series when selecting resampling methods by which to probe brain activity or identify functional networks in real-life fMRI experiments. We propose a statistical testing method to deal with the non-stationarity of continuous brain signals.
Assuntos
Mapeamento Encefálico , Imageamento por Ressonância Magnética , Encéfalo/diagnóstico por imagem , Mapeamento Encefálico/métodos , Análise de Fourier , Imageamento por Ressonância Magnética/métodosRESUMO
This study empirically assessed the strength and duration of short-term effects induced by brain reactions to closing/opening the eyes on a few well-known resting-state networks. We also examined the association between these reactions and subjects' cortisol levels. A total of 55 young adults underwent 8-min resting-state fMRI (rs-fMRI) scans under 4-min eyes-closed and 4-min eyes-open conditions. Saliva samples were collected from 25 of the 55 subjects before and after the fMRI sessions and assayed for cortisol levels. Our empirical results indicate that when the subjects were relaxed with their eyes closed, the effect of opening the eyes on conventional resting-state networks (e.g., default-mode, frontal-parietal, and saliency networks) lasted for roughly 60-s, during which we observed a short-term increase in activity in rs-fMRI time courses. Moreover, brain reactions to opening the eyes had a pronounced effect on time courses in the temporo-parietal lobes and limbic structures, both of which presented a prolonged decrease in activity. After controlling for demographic factors, we observed a significantly positive correlation between pre-scan cortisol levels and connectivity in the limbic structures under both conditions. Under the eyes-closed condition, the temporo-parietal lobes presented significant connectivity to limbic structures and a significantly positive correlation with pre-scan cortisol levels. Future research on rs-fMRI could consider the eyes-closed condition when probing resting-state connectivity and its neuroendocrine correlates, such as cortisol levels. It also appears that abrupt instructions to open the eyes while the subject is resting quietly with eyes closed could be used to probe brain reactivity to aversive stimuli in the ventral hippocampus and other limbic structures.
Assuntos
Mapeamento Encefálico , Hidrocortisona , Encéfalo/diagnóstico por imagem , Humanos , Imageamento por Ressonância Magnética/métodos , Descanso , Adulto JovemRESUMO
BACKGROUND: Previous studies in medical imaging have shown disparate abilities of artificial intelligence (AI) to detect a person's race, yet there is no known correlation for race on medical imaging that would be obvious to human experts when interpreting the images. We aimed to conduct a comprehensive evaluation of the ability of AI to recognise a patient's racial identity from medical images. METHODS: Using private (Emory CXR, Emory Chest CT, Emory Cervical Spine, and Emory Mammogram) and public (MIMIC-CXR, CheXpert, National Lung Cancer Screening Trial, RSNA Pulmonary Embolism CT, and Digital Hand Atlas) datasets, we evaluated, first, performance quantification of deep learning models in detecting race from medical images, including the ability of these models to generalise to external environments and across multiple imaging modalities. Second, we assessed possible confounding of anatomic and phenotypic population features by assessing the ability of these hypothesised confounders to detect race in isolation using regression models, and by re-evaluating the deep learning models by testing them on datasets stratified by these hypothesised confounding variables. Last, by exploring the effect of image corruptions on model performance, we investigated the underlying mechanism by which AI models can recognise race. FINDINGS: In our study, we show that standard AI deep learning models can be trained to predict race from medical images with high performance across multiple imaging modalities, which was sustained under external validation conditions (x-ray imaging [area under the receiver operating characteristics curve (AUC) range 0·91-0·99], CT chest imaging [0·87-0·96], and mammography [0·81]). We also showed that this detection is not due to proxies or imaging-related surrogate covariates for race (eg, performance of possible confounders: body-mass index [AUC 0·55], disease distribution [0·61], and breast density [0·61]). Finally, we provide evidence to show that the ability of AI deep learning models persisted over all anatomical regions and frequency spectrums of the images, suggesting the efforts to control this behaviour when it is undesirable will be challenging and demand further study. INTERPRETATION: The results from our study emphasise that the ability of AI deep learning models to predict self-reported race is itself not the issue of importance. However, our finding that AI can accurately predict self-reported race, even from corrupted, cropped, and noised medical images, often when clinical experts cannot, creates an enormous risk for all model deployments in medical imaging. FUNDING: National Institute of Biomedical Imaging and Bioengineering, MIDRC grant of National Institutes of Health, US National Science Foundation, National Library of Medicine of the National Institutes of Health, and Taiwan Ministry of Science and Technology.
Assuntos
Aprendizado Profundo , Neoplasias Pulmonares , Inteligência Artificial , Detecção Precoce de Câncer , Humanos , Estudos RetrospectivosRESUMO
BACKGROUND: Without timely diagnosis and treatment, tachycardia, also called tachyarrhythmia, can cause serious complications such as heart failure, cardiac arrest, and even death. The predictive performance of conventional clinical diagnostic procedures needs improvement in order to assist physicians in detecting risk early on. OBJECTIVE: We aimed to develop a deep tachycardia onset prediction (TOP-Net) model based on deep learning (ie, bidirectional long short-term memory) for early tachycardia diagnosis with easily accessible data. METHODS: TOP-Net leverages 2 easily accessible data sources: vital signs, including heart rate, respiratory rate, and blood oxygen saturation (SpO2) acquired continuously by wearable embedded systems, and electronic health records, containing age, gender, admission type, first care unit, and cardiovascular disease history. The model was trained with a large data set from an intensive care unit and then transferred to a real-world scenario in the general ward. In this study, 3 experiments incorporated merging patients' personal information, temporal memory, and different feature combinations. Six metrics (area under the receiver operating characteristic curve [AUROC], sensitivity, specificity, accuracy, F1 score, and precision) were used to evaluate predictive performance. RESULTS: TOP-Net outperformed the baseline models on the large critical care data set (AUROC 0.796, 95% CI 0.768-0.824; sensitivity 0.753, 95% CI 0.663-0.793; specificity 0.720, 95% CI 0.645-0.758; accuracy 0.721; F1 score 0.718; precision 0.686) when predicting tachycardia onset 6 hours in advance. When predicting tachycardia onset 2 hours in advance with data acquired from our hospital using the transferred TOP-Net, the 6 metrics were 0.965, 0.955, 0.881, 0.937, 0.793, and 0.680, respectively. The best performance was achieved using comprehensive vital signs (heart rate, respiratory rate, and SpO2) statistical information. CONCLUSIONS: TOP-Net is an early tachycardia prediction model that uses 8 types of data from wearable sensors and electronic health records. When validated in clinical scenarios, the model achieved a prediction performance that outperformed baseline models 0 to 6 hours before tachycardia onset in the intensive care unit and 2 hours before tachycardia onset in the general ward. Because of the model's implementation and use of easily accessible data from wearable sensors, the model can assist physicians with early discovery of patients at risk in general wards and houses.
RESUMO
Image-based teleconsultation using smartphones has become increasingly popular. In parallel, deep learning algorithms have been developed to detect radiological findings in chest X-rays (CXRs). However, the feasibility of using smartphones to automate this process has yet to be evaluated. This study developed a recalibration method to build deep learning models to detect radiological findings on CXR photographs. Two publicly available databases (MIMIC-CXR and CheXpert) were used to build the models, and four derivative datasets containing 6453 CXR photographs were collected to evaluate model performance. After recalibration, the model achieved areas under the receiver operating characteristic curve of 0.80 (95% confidence interval: 0.78-0.82), 0.88 (0.86-0.90), 0.81 (0.79-0.84), 0.79 (0.77-0.81), 0.84 (0.80-0.88), and 0.90 (0.88-0.92), respectively, for detecting cardiomegaly, edema, consolidation, atelectasis, pneumothorax, and pleural effusion. The recalibration strategy, respectively, recovered 84.9%, 83.5%, 53.2%, 57.8%, 69.9%, and 83.0% of performance losses of the uncalibrated model. We conclude that the recalibration method can transfer models from digital CXRs to CXR photographs, which is expected to help physicians' clinical works.
RESUMO
BCP1 is a protein enriched in the nucleus that is required for Mss4 nuclear export and identified as the chaperone of ribosomal protein Rpl23 in Saccharomyces cerevisiae. According to sequence homology, BCP1 is related to the mammalian BRCA2-interacting protein BCCIP and belongs to the BCIP protein family (PF13862) in the Pfam database. However, the BCIP family has no discernible similarity to proteins with known structure. Here, we report the crystal structure of BCP1, presenting an α/ß fold in which the central antiparallel ß-sheet is flanked by helices. Protein structural classification revealed that BCP1 has similarity to the GNAT superfamily but no conserved substrate-binding residues. Further modeling and protein-protein docking work provide a plausible model to explain the interaction between BCP1 and Rpl23. Our structural analysis presents the first structure of BCIP family and provides a foundation for understanding the molecular basis of BCP1 as a chaperone of Rpl23 for ribosome biosynthesis.
Assuntos
Proteínas Nucleares/química , Proteínas Nucleares/metabolismo , Proteínas de Saccharomyces cerevisiae/química , Proteínas de Saccharomyces cerevisiae/metabolismo , Saccharomyces cerevisiae/metabolismo , Sítios de Ligação/fisiologia , Cristalografia por Raios X/métodos , Conformação Proteica em Folha beta/fisiologia , Estrutura Secundária de Proteína/fisiologia , Proteínas Ribossômicas/química , Proteínas Ribossômicas/metabolismo , Ribossomos/metabolismoRESUMO
Plasma gasification of raw and torrefied woody, non-woody, and algal biomass using three different gasifying agents (air, steam, and CO2) is conducted through a thermodynamic analysis. The impacts of feedstock and reaction atmosphere on various performance indices such as syngas yield, pollutant emissions, plasma energy to syngas production ratio (PSR), and plasma gasification efficiency (PGE) are studied. Results show that CO2 plasma gasification gives the lowest PSR, thereby leading to the highest PGE among the three reaction atmospheres. Torrefied biomass displays increased syngas yield and PGE, but is more likely to have a negative environmental impact of N/S pollutants in comparison with raw one, especially for rice straw. However, the exception is for torrefied grape marc and macroalgae which produce lower amounts of S-species under steam and CO2 atmospheres. Overall, torrefied pine wood has the best performance for producing high quality syngas containing low impurities among the investigated feedstocks.