Prediction of Future Healthcare Expenses of Patients from Chest Radiographs Using Deep Learning: A Pilot Study

doi:10.21203/rs.3.rs-381448/v1

Download PDF

Research Article

Prediction of Future Healthcare Expenses of Patients from Chest Radiographs Using Deep Learning: A Pilot Study

https://doi.org/10.21203/rs.3.rs-381448/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Our objective was to develop deep learning models with chest radiograph data to predict healthcare costs and classify top-50% spenders. 21,872 frontal chest radiographs were retrospectively collected from 19,524 patients with at least 1-year spending data. Among the patients, 11,003 patients had 3 years of cost data, and 1678 patients had 5 years of cost data. Model performances were measured with area under the receiver operating characteristic curve (ROC-AUC) for classification of top-50% spenders and Spearman ρ for prediction of healthcare cost. The best model predicting 1-year (N=21,872) expenditure achieved ROC-AUC of 0.806 [95% CI, 0.793-0.819] for top-50% spender classification and ρ of 0.561 [0.536-0.586] for regression. Similarly, for predicting 3-year (N=12,395) expenditure, ROC-AUC of 0.771 [0.750-0.794] and ρ of 0.524 [0.489-0.559]; for predicting 5-year (N=1,779) expenditure ROC-AUC of 0.729 [0.667-0.729] and ρ of 0.424 [0.324-0.529]. Our deep learning model demonstrated the feasibility of predicting health care expenditure as well as classifying top 50% healthcare spenders at 1, 3, and 5 year(s), implying the feasibility of combining deep learning with information-rich imaging data to uncover hidden associations that may allude physicians. Such a model can be a starting point of making an accurate budget in reimbursement models in healthcare industries.

Biomedical Engineering

Health Economics & Outcomes Research

Future Healthcare Expenses

Chest Radiographs

Deep Learning

A Pilot Study

healthcare industries

Healthcare cost is an important barrier to healthcare access and account for a substantial fraction of the national budget. Health care expenses in a society are often distributed according to Pareto-like extreme value distributions, with top 50% spenders account for 97% of the total healthcare expenditures in the United States.^1,2 The top spenders tend to be patients who are sicker and often underserved in society.³ At an individual patient level, having a reliable cost estimation model can better prepare a patient financially and psychologically. At the hospital and national healthcare policy level, accurate identification and prediction of top healthcare spenders can be a starting point of making an accurate budget and planning appropriately for risk in reimbursement models based on health outcomes.

Recent advances in computer vision models, especially rapid advancements in convolutional neural networks (CNN), has contributed to a variety of applications.^5,6 Deep learning has especially been powerful in identifying mild to moderate associations that humans might not routinely predict or detect in dense images. For example, deep learning has been applied to imaging data to predict and diagnose various diseases such as pneumonia, Alzheimer’s disease, major thoracic diseases, and more.^7–10 In current practice, clinical radiologists might be able to extract information relevant to the clinical question related to the image data, but numerous mild to moderately powered hidden associations, whether clinically relevant or not, may exist within the dense imaging data of the radiograph.

We hypothesize that chest radiographs capture many general health indicators and thus may be utilized, potentially along with information on sex, age, and ZIP code, to predict future medical costs. The chest x-ray radiograph (CXR) is the most commonly performed radiological examination.⁴ It allows the examination of the heart, lung, airway, bones of the spine and chest, and blood vessels. Additionally, compared to text or categorical data, a radiograph is a high-dimensional and dense set of data that contains rich and unique information correlated to the patient, similar to that of a profile photo that can be revealing about a person. Ultimately, the prediction harnessed from chest radiographs may be used for proactive preventive care in order to potentially reduce the medical costs for high-risk groups. Therefore, this study aims to investigate the feasibility of developing deep learning models to predict healthcare costs and identify top 50% spenders.

Chest Radiograph Data

All procedures in this study were approved by the Institutional Review Board of University of California, San Francisco Medical Center, California, USA and performed in accordance with relevant guidelines and regulations. The institutional review board of University of California, San Francisco Medical Center, California, USA waived the need of informed consent for the retrospective use of the CXR data. All participants were non-obstetric adult patients who presented to the emergency department (ED) between July 1, 2012 and November 30, 2017 and received a chest radiograph at the ED or at an outpatient facility on the day of ED presentation. 34,743 frontal chest radiographs were initially identified belonging to 30,823 patients paired with corresponding patient’s age, sex, ZIP code, and cost at UCSF Medical Center within the consequent 1, 3, or 5 year(s). After inclusion and exclusion criteria are applied, 21,872 chest radiographs belonging to 19,524 patients with at least 1-year spending data were ultimately used in this study (Table 1). Patients ranged from 18 to 111 years old (56 ± 19.7 SD). Racial representation in our population was not significantly different from the census population in this geographic area (p = 0.650 in chi-square).²¹ Among the 19,524 patients, 11,003 patients had 3 years of cost data, and 1678 patients had 5 years of cost data (Table 2). At our institution, frontal chest radiographs for adult patients are typically obtained at 100–120 kVp with automatic exposure control. Source to detector distance is set at 72 inches unless modified for special reasons. Lateral chest radiographs were excluded. More than 90% of the images were acquired on GE and Philips hardware.

Table 1

Patient Demographics for Prediction of 1-Year Expenditure
Characteristics	Count (%)	# below ave cost (%)	# above ave cost (%)
Gender (p = 0.112)*
Female	9616 (49.3)	5091 (49.8)	4525 (48.7)
Male	9908 (50.7)	5133 (50.2)	4775 (51.3)
Age (p < 0.0001)*
18–37	4055 (20.8)	2949 (28.9)	1106 (11.9)
38–57	5423 (37.8)	3081 (30.1)	2342 (25.2)
58–76	6212 (31.8)	2634 (25.8)	3578 (38.5)
77–96	3639 (18.6)	1485 (14.5)	2154 (23.1)
97–116	195 (1.0)	75 (0.7)	120 (1.3)
Region (p < 0.0001)*
Bay Area	17153 (87.9)	9093 (88.9)	8060 (86.7)
Not Bay Area	2367 (12.1)	1129 (11.1)	1238 (13.3)
Not Assigned	4 (0.0)	2 (0.0)	2 (0.0)
Race (p < 0.0001)*
American Indian or Alaska Native	67 (0.3)	30 (0.3)	37 (0.4)
Asian	4196 (21.5)	2072 (20.3)	2124 (22.8)
Black or African American	2326 (11.9)	1219 (11.9)	1108 (11.9)
Native Hawaiian or Other Pacific Islander	454 (2.3)	285 (2.8)	169 (1.8)
Other	2923 (15.0)	1569 (15.3)	1354 (14.6)
Unknown/Declined	625 (3.2)	411 (4.0)	214 (2.3)
White or Caucasian	8933 (45.8)	4638 (45.4)	4295 (46.2)
*p-value is comparing the relationship between cost and the corresponding variables using chi-squared test for independence for categorical variables, two-tails at significance level 0.05.

Table 2

Performance Summary for Best Models on 1,3,5-Year(s) Expenditure.
Model	1 Year	3 Years	5 Years
Training Set Size	16,399	9,324	1,328
Classification ROC-AUC (95% CI); Corresponding Model	0.806 (0.793–0.819) Model TX1	0.771 (0.750–0.794) Model X	0.729 (0.667–0.729) Model TX2
Classification F1 (95% CI)	0.779 (0.766–0.791)	0.775 (0.756–0.794)	0.781 (0.736–0.826)
Regression Spearman ρ (95% CI); Corresponding Model	0.561 (0.536–0.586) Model TX2	0.524 (0.489–0.559) Model TX2	0.424 (0.324–0.529) Model X
Regression Pearson R (95% CI)	0.557 (0.532–0.581)	0.523 (0.487–0.558)	0.421 (0.314–0.530)
CI = confidence interval
*Pearson R is calculated using log₁₀-transformed data.

Data Exclusion Criterion

12,869 (37.0%) of 34,743 chest radiographs were excluded during data processing due to some missing patient information (Suppl. eFigure 1). 11,857 (92.1%) of the excluded radiographs did not have information available for their healthcare spending. The rest 1,012 (7.9%) of the excluded chest radiographs consisted of 8 with no associated sex, 128 with no associated ZIP codes and 876 whose ZIP code could not be matched to the median income. Pairwise chi-squared test was performed to account for statistical association between the 1,004 excluded CXRs (not including the 8 with unknown sex) and the 21,872 included CXRs, based on patient demographic variables such as sex, geographic area, and race. Results of two-sample t-test comparing the mean age between excluded and included CXRs is shown in Suppl. eTable 1. Among the 21,872 chest radiographs with 1 year expenditure, 9,477 (43.3%) are missing expenditure amounts for 3 years or longer and 20,073 (91.8%) are missing expenditure amounts for 5 years.

Healthcare Spending Data

Healthcare spending data was obtained from the cost accounting unit of the institution’s hospital financial department. Total healthcare spending was based on the sum of direct and indirect expenses attributed to patients’ hospital stay, pharmacy, laboratory, imaging, surgeries, and medical consultations over the time period during which they were included in the study. As an outcome, we selected total healthcare expenditure over the subsequent 1,3, and 5 years.

Model Architecture

Regression models were developed to predict healthcare expenditures, and binary classification models were developed to predict whether a participant’s healthcare expenditure was in the top 50%. Both regression and classification models were developed in four versions: (T) baseline model that relies only on patient sex, age, and ZIP code median income as input, (X) ResNet¹¹ with only CXR as input, (TX1) separately trained T and X model combined at final stage, and (TX2) modified ResNet trained end to end with CXR, age, sex, and per-ZIP code median income as input. The baseline (T) regression and classification baseline models were gradient boosting regressor ^12,13 and an AdaBoost Classifier ¹⁴ respectively implemented in the Python scikit-learn package with default parameters. The regression and classification CXR-only models (X) were a modified ResNet18 model and a modified ResNet50 model respectively. For combined model TX1 (Suppl. eFigure 2), the raw softmax score or final (regression) output from model (X) were concatenated to categorical data and then processed with model (T) approach to arrive at the output. For combined model TX2 (Suppl. eFigures 3), the neural network architectures from model (X) were modified at the final convolutional layers to allow the concatenation of the categorical data into the neural network model in an end-to-end fashion. See Supplemental eFigures 2–3 and eAppendix for implementation details.

Model Training and Evaluation

All versions of ResNet were initialized with weights pre-trained on ImageNet.^{15, 16} For all models, hyperparameters such as learning rate, linear layer dimension, number of linear layers, and others were empirically optimized via random search.¹⁷ After hyperparameter tuning and training, the models were evaluated against the pre-split test set.¹⁸ The training, validation, and test set were split by patient identification numbers to ensure that no two CXR from same patient is represented across multiple datasets. The outputs of the classification model were evaluated using the area under the receiver operator characteristic curve (ROC-AUC), and F1 score. The outputs of the regression model were measured using Pearson's R, and Spearman ρ. Confidence intervals (95%) were computed for all statistics. Each training and evaluation were performed for 1 year (21,872 CXR), 3 years (12,395 CXR) and 5 years of expenditure (1,779 CXR), respectively. Since 1-year expenditure data was the most complete, all following analysis should be assumed to be based on 1-year expenditure, unless mentioned otherwise.

Statistical Analysis

Pairwise chi-squared test was performed between the cost groups (above and below median expenditure patients) in order to inspect the relation with patient demographic variables such as age groups, sex, geographic area, and race. To examine the effect of reduction of dynamic ranges of costs on model performance, we trained and evaluated the baseline models on log₁₀ transformed costs and raw costs. The effects of variables sex, age, and race were analyzed on log₁₀-transformed costs. Supplemental eTable 2 contained the result of ANOVA analysis on four models. Model 1 is a baseline model with no factor and is used as a reference. Model 2 is a one-way ANOVA with factor sex. Model 3 is a two-way ANOVA with factors sex and age. Model 4 is a three-way ANOVA with factors sex, age, and race. For ROC-AUC, DeLong method¹⁹ was used to compare models pairwise. For error analysis, we interrogated whether the absolute difference between the true cost value and the predicted cost value is correlated with any of the patient demographic factors. The linear model used percentage differences (| true cost - predicted cost | / true cost) as the dependent variable and patient sex, race, age, median income, and overall true cost as the covariates.

Exploratory Data Analysis

The healthcare expenditures were log₁₀-transformed, making data distribution closer to a normal (Fig. 1A-B, Suppl. eFigure 4), with a median of 4.45 ($27,886), and mean ± SD of 4.43 ± 0.82 ($26,953). The ANOVA test across the demographic factors and healthcare expenditures revealed the following: race (p < 2e-16, Fig. 1C), median income (p = 1.23e-13, Fig. 1D), sex (p = 0.560, Fig. 1E) and age (p < 2e-16, Fig. 1F) were significantly associated with the variance in expenditures, except sex (p = 0.560). Higher median income was associated with lower healthcare expenditures (Pearson R=-0.0537, p = 5.4e-14; Spearman ρ=-0.0544, p = 2.4e-14; n = 19,612), an association that is robust to outliers and heavy tail effects (see Suppl. eAppendix).

Classification of Top-Spenders

Performance results of best models for each time period are shown in Fig. 2. Of the 4 classification models (T, X, TX1, and TX2) trained on 1-year expenditure data (n = 16,399 training CXR), the model TX1, leveraging both demographic data and features extracted by the CNN model, performed best with ROC-AUC of 0.806. For the classification within 3 years (n = 9,324), best results were achieved using model X, trained using only CXR as input, with ROC-AUC of 0.771. For the classification within 5 years (n = 1,328), best results were achieved using model TX2 (ROC-AUC of 0.729) that combined the inputs of T and X trained end-to-end. Pairwise comparison of the 1-year expenditure results (Suppl. eFigure 5) showed that the ROC-AUC of the classification model T is significantly different from all other models (p < 2.2e-16), unlike differences between X and TX1, TX1 and TX2 models (p-values > 0.05). We use the TX1 model for subsequent 1-year expenditure analysis.

Prediction of Healthcare Costs

The predicted costs within 1, 3, and 5 year(s) were demonstrated to have a correlation with true costs as illustrated in the joint histogram (Fig. 3A). The prediction model for 1-year costs ourperformed the other models for 3- and 5-year costs. The best model for 1-year cost is TX2 that achieved Spearman ρ of 0.561 (p = 2.362e-271) (Fig. 3B). 3-years costs follow with Spearman ρ of 0.524 (p = 5.423e-129) by model TX2. 5-years costs come last with Spearman ρ of 0.424 (p = 3.915e-13) by model X. Examples of chest radiographs pairing with Grad-CAM²⁰ maps of the chest radiographs predicted using 1-year expenditure data are shown in Fig. 4.

Missing Data and Error Analysis

Missingness in our dataset was correlated with sex, geographic area, and race (p < 0.001 in chi-square), as well as age (p < 0.001, two-sample t-test, Suppl. eTable 1). Only overall cost (adjusted R² = 0.014, p < 0.01) and ZIP code median income (adjusted R² = 0.040, p < 0.01) were associated with residuals of prediction for model TX1.

We demonstrated the feasibility of predicting healthcare costs and classifying top-50% spenders by using deep learning models based on chest radiographs (CXR) that are widely available in clinics and hospitals. The models were developed to identify patients who are likely to incur high healthcare expenditure and predict their subsequent amount of healthcare spending within 1, 3, and 5 years. Unlike physicians who are trained to identify only a handful of imaging biomarkers known to medical literature, our deep learning algorithm is able to take into account thousands of imaging features of weak to moderate correlations with healthcare spending as presented in the training set. When a CXR is evaluated by the deep learning algorithm, its pixels are aggregated, transformed, and passed through many layers of filters with each layer extracting different lines, angles, patterns, and associations. As those extracted features are then passed upstream to higher-level filters, they are compared to the thousands of CXR that the algorithm was trained on. All these numbers finally converge to the estimated cost.^21,22 Considering that CXR tend to be standardized, deep learning algorithms are trained to be extremely sensitive to details that clinical radiologists may not typically recognize.

From a data scientist perspective, the ability of deep learning algorithms to predict healthcare expenditure from CXR is a testament to the vast amounts of information hidden in imaging data that can be leveraged with data science. The addition of other demographic and clinical variables to the imaging data resulted in minimal improvements to the model, despite the baseline models showing that sex, age, and zip code median income are individually associated with healthcare expenditures. This again affirms the presence of rich information within imaging data and the ability of deep learning models to extract them. It is important to note that deep learning algorithms are, at large, approximations based on a large volume of data.²³ The causality for each prediction cannot be definitely deduced.²⁴ As of any machine learning predictions, it cannot be used as definitive proof of a patient’s health or future health expenditure. In addition, there remains ethical concerns as well if the algorithm is used to deny coverage by insurance companies. Nevertheless, the deep learning algorithm can be potentially used by government or insurance companies to identify high-risk individuals and take appropriate actions to secure their health and reduce cost. Such predictions can provide an important starting point in identifying high risk patients to achieve reduction in their healthcare spending and encouraging lifestyle modifications and more intensive medical management to achieve better medical and financial outcomes.

From a clinical perspective, the deep learning algorithm takes into account a combination of demographic factors (age, sex), baseline health factors (weight, bone health), as well as clinical diseases (e.g., enlarged heart, osteophytes, etc) that are inferred from CXR. For example, having hemodialysis access or enlarged heart from congestive heart failure could be strong indicators of higher healthcare spending predicted by the algorithm. Having replaced hardware or numerous osteophytes could be indicative of older age, which in itself is a predictor of higher healthcare spending as well. While the algorithm does not explicitly give these medical diagnoses when it arrives at its final spending prediction, the algorithm is able to incorporate numerous weak to moderately associated cost predictors in the CXR and assemble them into the final cost predictions. Our algorithm could be used in outpatient settings to estimate approximate future healthcare costs such that patients, doctors, and insurance companies would have a reliable indicator to consider when making patient treatment and financial decisions. The identified high-risk patients could be subject to more intensive preventive medical interventions and close follow-up visits to modify patient outcomes. The algorithm could also be used to identify patients with CXR that appear normal according to current clinical radiological standards but are still at risk for high medical costs. Similar to most deep learning algorithms, the application of ours can potentially be automatic, fast, scalable, and relatively low cost when compared to other services in the healthcare system.

Several limitations to the study should be noted, primarily related to selection bias inherent in this particular dataset. First, the performance differences between 1, 3, and 5-year models were observed, which can be attributed to both drastic differences in sample size as well as inherent loss of predictive information about the future. The 3-year expenditure model performed slightly worse than the 1-year expenditure model using 56.7% of the sample size. The 5-year model used only 8.1% of the sample size but still achieved reasonably accurate classification and regression results. We believe that with more data the 5-year expenditure model would show even more promise. Second, the development and testing of the model involved data originating from a single hospital system and most lived in the San Francisco Bay Area in the United States healthcare system. The model will likely not generalize to the non-American healthcare system due to the particular structure of healthcare expenses. However, a similar approach can be undertaken to build a new model with any local dataset. Third, missing data (mainly due to missing financial information) constituted 37% of the originally extracted dataset and they were missing not at random. For example, homeless patients may not have had a zip code available. Fourth, patient death information was not available and could have had a variable impact on healthcare costs. Fifth, the dataset did not include inpatient cases and portable CXR.

We demonstrated the potential of deep learning algorithms to predict 1,3, and 5-years patient healthcare expenditure based on a frontal chest radiograph even in the absence of additional clinical information. This study confirms that radiological imaging indeed contains rich information that may not be routinely extracted by human radiologists but can be analyzed by the power of big data and deep learning. Successfully predicting healthcare expenditure can potentially be an important first step towards improving health policy and medical interventions to address patient care and societal costs.

Acknowledgment

Research reported in this publication was supported in part by NIBIB T32 grant T32EB001631 (JHS), National Cancer Institute grant UH2CA203792, National Library of Medicine grant U01LM012675 (DH), National Heart, Lung, and Blood Institute grant R01HL135490, and National Institute of Biomedical Imaging and Bioengineering grant R01EB026331 (YS).

AUTHOR CONTRIBUTIONS STATEMENT

All authors have made an intellectual contribution to the manuscript and have agreed to the submission. We request that authors JHS and YC be co-first authors with equal contribution as both have been extensively involved in project design, model development, and writing of the manuscript. DL, TV, BLF, and YS contributed to the data collection, project design, and manuscript editing. JY and KO, DH contributed to the project design and manuscript editing.

ADDITIONAL INFORMATION

None of the authors reported a relevant conflict of interest.

Steven B. Cohen, William Yu. The Concentration and Persistence in the Level of Health Expenditures over Time: Estimates for the U.S. Population, 2008–2009. Rockville, MD: Agency for Healthcare Research and Quality; 2012. https://meps.ahrq.gov/data_files/publications/st354/stat354.shtml. Accessed August 3, 2019.
Bradley Sawyer, Gary Claxton. How do health expenditures vary across the population? Peterson-Kais Health Syst Tracker. January 2019. https://www.healthsystemtracker.org/chart-collection/health-expenditures-vary-across-population/. Accessed August 3, 2019.
High Out‐of‐Pocket Medical Spending among the Poor and Elderly in Nine Developed Countries. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4946036/. Accessed December 15, 2019.
Speets AM, van der Graaf Y, Hoes AW, et al. Chest radiography in general practice: indications, diagnostic yield and consequences for patient management. Br J Gen Pract. 2006;56(529):574-578.
Greenspan H, Ginneken B van, Summers RM. Guest Editorial Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique. IEEE Trans Med Imaging. 2016;35(5):1153-1159. doi:10.1109/TMI.2016.2553401
Kermany DS, Goldbaum M, Cai W, et al. Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning. Cell. 2018;172(5):1122-1131.e9. doi:10.1016/j.cell.2018.02.010
Rajpurkar P, Irvin J, Zhu K, et al. CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning. ArXiv171105225 Cs Stat. November 2017. http://arxiv.org/abs/1711.05225. Accessed August 6, 2019.
Lu MT, Ivanov A, Mayrhofer T, Hosny A, Aerts HJWL, Hoffmann U. Deep Learning to Assess Long-term Mortality From Chest Radiographs. JAMA Netw Open. 2019;2(7):e197416-e197416. doi:10.1001/jamanetworkopen.2019.7416
Pasa F, Golkov V, Pfeiffer F, Cremers D, Pfeiffer D. Efficient Deep Network Architectures for Fast Chest X-Ray Tuberculosis Screening and Visualization. Sci Rep. 2019;9(1):1-9. doi:10.1038/s41598-019-42557-4
Hwang EJ, Park S, Jin K-N, et al. Development and Validation of a Deep Learning–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs. JAMA Netw Open. 2019;2(3):e191095-e191095. doi:10.1001/jamanetworkopen.2019.1095
He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. ArXiv151203385 Cs. December 2015. http://arxiv.org/abs/1512.03385. Accessed June 5, 2019.
Friedman JH. Greedy Function Approximation: A Gradient Boosting Machine. Ann Stat. 2001;29(5):1189-1232.
Friedman JH. Stochastic gradient boosting. Comput Stat Data Anal. 2002;38(4):367-378. doi:10.1016/S0167-9473(01)00065-2
Freund Y, Schapire RE. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J Comput Syst Sci. 1997;55(1):119-139. doi:10.1006/jcss.1997.1504
Deng J, Dong W, Socher R, Li L, Kai Li, Li Fei-Fei. ImageNet: A large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition. ; 2009:248-255. doi:10.1109/CVPR.2009.5206848
Bengio Y. Deep Learning of Representations for Unsupervised and Transfer Learning. In: Proceedings of ICML Workshop on Unsupervised and Transfer Learning. Vol 27. Washington, USA; 2012:17-37. http://proceedings.mlr.press/v27/bengio12a.html. Accessed July 11, 2019.
Bergstra J, Bengio Y. Random Search for Hyper-Parameter Optimization. J Mach Learn Res. 2012;13(Feb):281-305.
Parmar C, Barry JD, Hosny A, Quackenbush J, Aerts HJWL. Data Analysis Strategies in Medical Imaging. Clin Cancer Res Off J Am Assoc Cancer Res. 2018;24(15):3492-3499. doi:10.1158/1078-0432.CCR-18-0385
DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837-845.
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization. 2017 IEEE International Conference on Computer Vision (ICCV). doi: 10.1109/iccv.2017.74.
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88. doi:10.1016/j.media.2017.07.005
Shen D, Wu G, Suk H-I. Deep Learning in Medical Image Analysis. Annu Rev Biomed Eng. 2017;19(1):221-248. doi:10.1146/annurev-bioeng-071516-044442
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436-444. doi:10.1038/nature14539
Samek W, Wiegand T, Müller K-R. Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models. ArXiv170808296 CsAI. August 2017. http://arxiv.org/abs/1708.08296. Accessed August 6, 2019.

No competing interests reported.

ScientificReportCXRcost0401supplementfixed.docx

Download PDF

Editorial decision: Major revision
30 Sep, 2021
Reviews received at journal
28 Sep, 2021
Reviewers agreed at journal
05 Sep, 2021
Reviews received at journal
29 May, 2021
Reviewers agreed at journal
13 May, 2021
Reviewers invited by journal
13 May, 2021
Editor assigned by journal
13 May, 2021
Editor invited by journal
03 Apr, 2021
Submission checks completed at journal
03 Apr, 2021
First submitted to journal
31 Mar, 2021

You are reading this latest preprint version

Prediction of Future Healthcare Expenses of Patients from Chest Radiographs Using Deep Learning: A Pilot Study

Status:

Version 1

Abstract

Figures

Introduction

Methods

Chest Radiograph Data

Data Exclusion Criterion

Healthcare Spending Data

Model Architecture

Model Training and Evaluation

Statistical Analysis

Results

Exploratory Data Analysis

Classification of Top-Spenders

Prediction of Healthcare Costs

Missing Data and Error Analysis

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1