In healthcare, artificial intelligence (AI) technologies have the potential to create significant value by improving time-sensitive outcomes while lowering error rates for each patient. Diagnostic images, clinical notes, and reports are increasingly generated and stored in electronic medical records. This heterogeneous data presenting us with challenges in data analytics and reusability that is by nature has high complexity, thereby necessitating novel ways to store, manage and process, and reuse big data. This presents an urgent need to develop new, scalable, and expandable AI infrastructure and analytical methods that can enable healthcare providers to access knowledge for individual patients, yielding better decisions and outcomes. In this review article, we briefly discuss the nature of data in breast cancer study and the role of AI for generating "smart data" which offer actionable information that supports the better decision for personalized medicine for individual patients. In our view, the biggest challenge is to create a system that makes data robust and smart for healthcare providers and patients that can lead to more effective clinical decision-making, improved health outcomes, and ultimately, managing the healthcare outcomes and costs. We highlight some of the challenges in using breast cancer data and propose the need for an AI-driven environment to address them. We illustrate our vision with practical use cases and discuss a path for empowering the study of breast cancer databases with the application of AI and future directions.

The study of precision medicine that measures the effects of social, cultural, and environmental influences on health is essential to improve health outcomes. Race is a social concept used historically to divide, track, control populations, and reinforce social hierarchies. Beyond genetics, race is also a surrogate for other socioeconomic factors affecting patient outcomes. Our data analytics study aims to analyze the Electronic Medical Record (EMR) to study patients of different races in diagnosing and treating Coronary Artery Disease (CAD). We found no race discrepancies at the University of California San Francisco Medical Centers. This study opens several new hypotheses for further research in this crucial field.

Melanoma will affect an estimated 73,000 new cases this year and result in 9,000 deaths, yet precise diagnosis remains a serious problem. Without early detection and preventative care, melanoma can quickly spread to become fatal (Stage IV 5-year survival rate is 20-10%) from a once localized skin lesion (Stage IA 5- year survival rate is 97%). There is no biomarker for melanoma in clinical use, and the current diagnostic criteria for skin lesions remains subjective and imprecise. Accurate diagnosis of melanoma relies on a histopathologic gold standard; thus, aggressive excision of melanocytic skin lesions has been the mainstay of treatment. It is estimated that 36 biopsies are performed for every melanoma confirmed by pathology among excised lesions. There is significant morbidity in misdiagnosing melanoma such as progression of the disease for a false negative prediction vs the risks of unnecessary surgery for a false positive prediction. Every year, poor diagnostic precision adds an estimated $673 million in overall cost to manage the disease. Currently, manual dermatoscopic imaging is the standard of care in selecting atypical skin lesions for biopsy, and at best it achieves 90% sensitivity but only 59% specificity when performed by an expert dermatologist. Many computer vision (CV) algorithms perform better than dermatologists in classifying skin lesions although not significantly so in clinical practice. Meanwhile, open source deep learning (DL) techniques in CV have been gaining dominance since 2012 for image classification, and today DL can outperform humans in classifying millions of digital images with less than 5% error rates. Moreover, DL algorithms are readily run on commoditized hardware and have a strong online community of developers supporting their rapid adoption. In this work, we performed a successful pilot study to show proof of concept to DL skin pathology from images. However, DL algorithms must be trained on very large labelled datasets of images to achieve high accuracy. Here, we begin to assemble a large imageset of skin lesions from the UCSF and the San Francisco Veterans Affairs Medical Center (VAMC) dermatology clinics that are well characterized by their underlying pathology, on which to train DL algorithms. If trained on sufficient data, we hypothesize that our approach will significantly outperform general dermatologists in predicting skin lesion pathology. We posit that our work will allow for precision diagnosis of melanoma from widely available digital photography, which may optimize the management of the disease by decreasing unnecessary office visits and the significant morbidity and cost of melanoma misdiagnosis.

Study design

A retrospective study at a single academic institution.

Objective

The purpose of this study is to utilize machine learning to predict hospital length of stay (LOS) and discharge disposition following adult elective spine surgery, and to compare performance metrics of machine learning models to the American College of Surgeon's National Surgical Quality Improvement Program's (ACS NSQIP) prediction calculator.

Summary of background data

A total of 3678 adult patients undergoing elective spine surgery between 2014 and 2019, acquired from the electronic health record.

Methods

Patients were divided into three stratified cohorts: cervical degenerative, lumbar degenerative, and adult spinal deformity groups. Predictive variables included demographics, body mass index, surgical region, surgical invasiveness, surgical approach, and comorbidities. Regression, classification trees, and least absolute shrinkage and selection operator (LASSO) were used to build predictive models. Validation of the models was conducted on 16% of patients (N=587), using area under the receiver operator curve (AUROC), sensitivity, specificity, and correlation. Patient data were manually entered into the ACS NSQIP online risk calculator to compare performance. Outcome variables were discharge disposition (home vs. rehabilitation) and LOS (days).

Results

Of 3678 patients analyzed, 51.4% were male (n=1890) and 48.6% were female (n=1788). The average LOS was 3.66 days. In all, 78% were discharged home and 22% discharged to rehabilitation. Compared with NSQIP (Pearson R2 =0.16), the predictions of poisson regression ( R2 =0.29) and LASSO ( R2 =0.29) models were significantly more correlated with observed LOS ( P =0.025 and 0.004, respectively). Of the models generated to predict discharge location, logistic regression yielded an AUROC of 0.79, which was statistically equivalent to the AUROC of 0.75 for NSQIP ( P =0.135).

Conclusion

The predictive models developed in this study can enable accurate preoperative estimation of LOS and risk of rehabilitation discharge for adult patients undergoing elective spine surgery. The demonstrated models exhibited better performance than NSQIP for prediction of LOS and equivalent performance to NSQIP for prediction of discharge location.

Applying state-of-the-art machine learning techniques to medical images requires a thorough selection and normalization of input data. One of such steps in digital mammography screening for breast cancer is the labeling and removal of special diagnostic views, in which diagnostic tools or magnification are applied to assist in assessment of suspicious initial findings. As a common task in medical informatics is prediction of disease and its stage, these special diagnostic views, which are only enriched among the cohort of diseased cases, will bias machine learning disease predictions. In order to automate this process, here, we develop a machine learning pipeline that utilizes both DICOM headers and images to predict such views in an automatic manner, allowing for their removal and the generation of unbiased datasets. We achieve AUC of 99.72% in predicting special mammogram views when combining both types of models. Finally, we apply these models to clean up a dataset of about 772,000 images with expected sensitivity of 99.0%. The pipeline presented in this paper can be applied to other datasets to obtain high-quality image sets suitable to train algorithms for disease detection.

Background

Women continue to have worse Coronary Artery Disease (CAD) outcomes than men. The causes of this discrepancy have yet to be fully elucidated. The main objective of this study is to detect gender discrepancies in the diagnosis and treatment of CAD.

Methods

We used data analytics to risk stratify ~32,000 patients with CAD of the total 960,129 patients treated at the UCSF Medical Center over an 8 year period. We implemented a multidimensional data analytics framework to trace patients from admission through treatment to create a path of events. Events are any medications or noninvasive and invasive procedures. The time between events for a similar set of paths was calculated. Then, the average waiting time for each step of the treatment was calculated. Finally, we applied statistical analysis to determine differences in time between diagnosis and treatment steps for men and women.

Results

There is a significant time difference from the first time of admission to diagnostic Cardiac Catheterization between genders (p-value = 0.000119), while the time difference from diagnostic Cardiac Catheterization to CABG is not statistically significant.

Conclusion

Women had a significantly longer interval between their first physician encounter indicative of CAD and their first diagnostic cardiac catheterization compared to men. Avoiding this delay in diagnosis may provide more timely treatment and a better outcome for patients at risk. Finally, we conclude by discussing the impact of the study on improving patient care with early detection and managing individual patients at risk of rapid progression of CAD.

Early detection plays a key role to enhance the outcome for Coronary Artery Disease. We utilized a big data analytics platform on ∼32,000 patients to trace patients from the first encounter to CAD treatment. There are significant gender-based differences in patients younger than 60 from the time of the first encounter to Coronary Artery Bypass Grafting with a p-value=0.03. This recognition makes significant changes in outcome by avoiding delay in treatment.

The development of an ontology facilitates the organization of the variety of concepts used to describe different terms in different resources. The proposed ontology will facilitate the study of cardiothoracic surgical education and data analytics in electronic medical records (EMR) with the standard vocabulary.

The identification of novel disease associations using big-data for patient care has had limited success. In this study, we created a longitudinal disease network of traced readmissions (disease trajectories), merging data from over 10.4 million inpatients through the Healthcare Cost and Utilization Project, which allowed the representation of disease progression mapping over 300 diseases. From these disease trajectories, we discovered an interesting association between schizophrenia and rhabdomyolysis, a rare muscle disease (incidence < 1E-04) (relative risk, 2.21 [1.80-2.71, confidence interval = 0.95], P-value 9.54E-15). We validated this association by using independent electronic medical records from over 830,000 patients at the University of California, San Francisco (UCSF) medical center. A case review of 29 rhabdomyolysis incidents in schizophrenia patients at UCSF demonstrated that 62% are idiopathic, without the use of any drug known to lead to this adverse event, suggesting a warning to physicians to watch for this unexpected risk of schizophrenia. Large-scale analysis of disease trajectories can help physicians understand potential sequential events in their patients.

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 = 1779) 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 to physicians. Such a model can be a starting point of making an accurate budget in reimbursement models in healthcare industries.