Search

To assess women’s knowledge, patient experience, and treatment decision making regarding overactive bladder (OAB) using digital ethnography. Online posts were identified using a data mining service. Two hundred randomized posts were reviewed and coded using grounded theory. We then applied a latent Dirichlet allocation (LDA) probabilistic topic modeling process to review the entire collection of identified posts. A total of 2618 posts by 1867 unique users from 203 different websites were identified. Our analysis yielded six themes: the impact of OAB on quality of life, patient-physician interactions, online engagement, symptom management, patient knowledge acquisition, and alternative therapies. Overall, online communities are a source of support for women to self-manage the OAB symptom complex and help overcome treatment pathway challenges. Digital ethnography provides insight into patient knowledge and barriers to patient-centered care, which are important to improve patient outreach. Additionally, we identify similar findings to prior work, indicating the reliability of studying social media.

Objective To demonstrate enabling multi-institutional training without centralizing or sharing the underlying physical data via federated learning (FL). Materials and Methods Deep learning models were trained at each participating institution using local clinical data, and an additional model was trained using FL across all of the institutions. Results We found that the FL model exhibited superior performance and generalizability to the models trained at single institutions, with an overall performance level that was significantly better than that of any of the institutional models alone when evaluated on held-out test sets from each institution and an outside challenge dataset. Discussion The power of FL was successfully demonstrated across 3 academic institutions while avoiding the privacy risk associated with the transfer and pooling of patient data. Conclusion Federated learning is an effective methodology that merits further study to enable accelerated development of models across institutions, enabling greater generalizability in clinical use.

Diagnostic radiology (DxR), having had successful serial co-evolutions with imaging equipment and PACS, is faced with another. With a backdrop termed "globotics transition," it should create an IT and informatics infrastructure capable of integrating artificial intelligence (AI) into current critical communication functions of PACS and incorporating functions currently residing in balkanized products. DxR will face the challenge of adopting sustaining and disruptive AI innovations simultaneously. In this co-evolution, a major selection force for AI will be increasing the flow of information and patients; "increasing" means faster flow over larger areas defined by geography and content. Larger content includes a broader spectrum of imaging and nonimaging information streams that facilitate medical decision making. Evolution to faster flow will gravitate toward a hierarchical IT architecture consisting of many small channels feeding into fewer larger channels, something potentially difficult for current PACS. Smartphone-like architecture optimized for communication and integration could provide a large-channel backbone and many smaller feeding channels for basic functions, as well as those needing to innovate rapidly. New, more flexible architectures stimulate market competition in which DxR could act as an artificial selection force to influence development of faster increased flow in current PACS companies, in disruptors such as consolidated AI companies, or in entirely new entrants like Apple or Google. In this co-evolution, DxR should be able to stimulate design of a modern communication medium that increases the flow of information and decreases the time and energy necessary to absorb it, thereby creating even more indispensable clinical value for itself.

OBJECTIVE To detail the development of an electronic report that graphically conveys all relevant information from targeted prostate biopsy. METHODS The Urology Integrated Diagnostic Report (Uro-IDR) is based on a published framework (RadPath) which enables the compilation of diagnostic data from urology, radiology, and pathology. Each component of the Uro-IDR is generated by the contributing clinician, is assembled in one document, and provides correlation of the 3 inputs at a glance. Upon completion, the Uro-IDR is automatically linked to the electronic medical record as an interactive file and can also be downloaded for offline sharing as a PDF. RESULTS At our institution, 1638 individual Uro-IDRs were generated between June 2016 and April 2019. There were 5715 views of these documents via the EMR. The average turnaround time for the creation of an individual report decreased from nearly 8 days at the time of its launch to 2 days after 6 months of use. The average time for report generation was 22 seconds for the pathologist and 69 seconds for the radiologist. An instructive video is linked to this article. CONCLUSION The Uro-IDR has proven to be a feasible, efficient, clinically useful form to concisely transmit key information about targeted prostate biopsy to both clinicians and patients.

Author(s): Doepke, Laura; Arnold, Corey; Hoyt, Anne; Enzmann, Dieter | Abstract: Most breast imaging practices have an organized system for reviewing the results of breast biopsies, assessing radiologic-histologic concordance, and making follow-up management recommendations. Yet, historically, the breast pathology report and radiologic addendum have been separated within the electronic medical record. Having the two entities fused into one integrated diagnostic report (IDR) has provided both a concise summary of the pertinent findings of imaging and histologic examinations related to the patient’s care and clear guidance for treatment and follow-up. Direct correlation between imaging and histologic findings has been shown to decrease discordance between these findings and increase diagnostic accuracy. The IDR is also a useful summary for conferences of tumor boards and multidisciplinary clinics. In addition, the software that is used for generating the IDR is capable of identifying patients who may benefit from clinical trials.

Mechanical thrombectomy (MTB) is one of the two standard treatment options for Acute Ischemic Stroke (AIS) patients. Current clinical guidelines instruct the use of pretreatment imaging to characterize a patient's cerebrovascular flow, as there are many factors that may underlie a patient's successful response to treatment. There is a critical need to leverage pretreatment imaging, taken at admission, to guide potential treatment avenues in an automated fashion. The aim of this study is to develop and validate a fully automated machine learning algorithm to predict the final modified thrombolysis in cerebral infarction (mTICI) score following MTB. A total 321 radiomics features were computed from segmented pretreatment MRI scans for 141 patients. Successful recanalization was defined as mTICI score >= 2c. Different feature selection methods and classification models were examined in this study. Our best performance model achieved 74.42±2.52% AUC, 75.56±4.44% sensitivity, and 76.75±4.55% specificity, showing a good prediction of reperfusion quality using pretreatment MRI. Results suggest that MR images can be informative to predicting patient response to MTB, and further validation with a larger cohort can determine the clinical utility.

PURPOSE: The appropriate number of systematic biopsy cores to retrieve during magnetic resonance imaging (MRI)-targeted prostate biopsy is not well defined. We aimed to demonstrate a biopsy sampling approach that reduces required core count while maintaining diagnostic performance. MATERIALS AND METHODS: We collected data from a cohort of 971 men who underwent MRI-ultrasound fusion targeted biopsy for suspected prostate cancer. A regional targeted biopsy (RTB) was evaluated retrospectively; only cores within 2 cm of the margin of a radiologist-defined region of interest were considered part of the RTB. We compared detection rates for clinically significant prostate cancer (csPCa) and cancer upgrading rate on final whole mount pathology after prostatectomy between RTB, combined, MRI-targeted, and systematic biopsy. RESULTS: A total of 16,459 total cores from 971 men were included in the study data sets, of which 1,535 (9%) contained csPCa. The csPCa detection rates for systematic, MRI-targeted, combined, and RTB were 27.0% (262/971), 38.3% (372/971), 44.8% (435/971), and 44.0% (427/971), respectively. Combined biopsy detected significantly more csPCa than systematic and MRI-targeted biopsy (p

Recent developments in machine learning algorithms have enabled models to exhibit impressive performance in healthcare tasks using electronic health record (EHR) data. However, the heterogeneous nature and sparsity of EHR data remains challenging. In this work, we present a model that utilizes heterogeneous data and addresses sparsity by representing diagnoses, procedures, and medication codes with temporal Hierarchical Clinical Embeddings combined with Topic modeling (HCET) on clinical notes. HCET aggregates various categories of EHR data and learns inherent structure based on hospital visits for an individual patient. We demonstrate the potential of the approach in the task of predicting depression at various time points prior to a clinical diagnosis. We found that HCET outperformed all baseline methods with a highest improvement of 0.07 in precision-recall area under the curve (PRAUC). Furthermore, applying attention weights across EHR data modalities significantly improved the performance as well as the model's interpretability by revealing the relative weight for each data modality. Our results demonstrate the model's ability to utilize heterogeneous EHR information to predict depression, which may have future implications for screening and early detection.

Histopathological images provide rich information for disease diagnosis. Large numbers of histopathological images have been digitized into high resolution whole slide images, opening opportunities in developing computational image analysis tools to reduce pathologists' workload and potentially improve inter- and intra- observer agreement. Most previous work on whole slide image analysis has focused on classification or segmentation of small pre-selected regions-of-interest, which requires fine-grained annotation and is non-trivial to extend for large-scale whole slide analysis. In this paper, we proposed a multi-resolution multiple instance learning model that leverages saliency maps to detect suspicious regions for fine-grained grade prediction. Instead of relying on expensive region- or pixel-level annotations, our model can be trained end-to-end with only slide-level labels. The model is developed on a large-scale prostate biopsy dataset containing 20,229 slides from 830 patients. The model achieved 92.7% accuracy, 81.8% Cohen's Kappa for benign, low grade (i.e. Grade group 1) and high grade (i.e. Grade group >= 2) prediction, an area under the receiver operating characteristic curve (AUROC) of 98.2% and an average precision (AP) of 97.4% for differentiating malignant and benign slides. The model obtained an AUROC of 99.4% and an AP of 99.8% for cancer detection on an external dataset., 9 pages, 6 figures

Objective To characterize the decision-making process and illness experience of women with pelvic organ prolapse (POP) using large-scale social media analysis. Methods Digital ethnographic analysis of online posts identified through data mining was performed. Grounded theory methodology was applied to 200 posts via traditional hand coding. To supplement our qualitative approach, we applied a Latent Dirichlet Allocation probabilistic topic modeling process to review the entire data set of identified posts to ensure thematic saturation. Results There were 3451 posts by 2088 unique users from 117 websites worldwide that were identified via social media data mining. We found that the anonymity of online forums allowed for information and support exchange among women with POP. Our analysis revealed that the exchange of online information aids in the decision-making process and, in some instances, appears to be the primary source of information. There was confusion about the anatomical and surgical complexities of prolapse. Our study also identified misconceptions, perceived risk factors, prevention methods, and management recommendations that were discussed online. Conclusions This large-scale online community-based analysis demonstrated the utility of social media to better understand women's experiences with POP. Thematic findings highlighted essential concerns and challenges involved in the surgical decision-making process and the understating of the anatomical complexity of sector defects, specifically to cystocele, rectocele, State specific defects.

Purpose Developing large-scale datasets with research-quality annotations is challenging due to the high cost of refining clinically generated markup into high precision annotations. We evaluated the direct use of a large dataset with only clinically generated annotations in development of high-performance segmentation models for small research-quality challenge datasets. Materials and methods We used a large retrospective dataset from our institution comprised of 1,620 clinically generated segmentations, and two challenge datasets (PROMISE12: 50 patients, ProstateX-2: 99 patients). We trained a 3D U-Net convolutional neural network (CNN) segmentation model using our entire dataset, and used that model as a template to train models on the challenge datasets. We also trained versions of the template model using ablated proportions of our dataset, and evaluated the relative benefit of those templates for the final models. Finally, we trained a version of the template model using an out-of-domain brain cancer dataset, and evaluated the relevant benefit of that template for the final models. We used five-fold cross-validation (CV) for all training and evaluation across our entire dataset. Results Our model achieves state-of-the-art performance on our large dataset (mean overall Dice 0.916, average Hausdorff distance 0.135 across CV folds). Using this model as a pre-trained template for refining on two external datasets significantly enhanced performance (30% and 49% enhancement in Dice scores respectively). Mean overall Dice and mean average Hausdorff distance were 0.912 and 0.15 for the ProstateX-2 dataset, and 0.852 and 0.581 for the PROMISE12 dataset. Using even small quantities of data to train the template enhanced performance, with significant improvements using 5% or more of the data. Conclusion We trained a state-of-the-art model using unrefined clinical prostate annotations and found that its use as a template model significantly improved performance in other prostate segmentation tasks, even when trained with only 5% of the original dataset.

Learning under deep neural network models has shown interesting abilities to generate semi-structured sequences. Such generated fragments can appear to be remarkably like real data, yet applications of these generative techniques are generally lacking. This is partly due to the inability of recurrent neural network sequences to be semantically connected amongst long related fragments. One way to increase the semantics of these fragments is to combine features learned from topic modeling with those of recurrent neural networks. This combination ensures a higher correlation between generated sequences, much like real data. In this paper we investigate existing approaches to adding topics into the recurrent neural network. We then develop a novel approach called Topic-RNN, which represents the state-of-the-art technique of integrating topics into the RNN. Our results show this method outperforms existing approaches substantially. Additionally, we show that in such cases where topic information is available, Topic-RNN surpasses word level RNNs for the task of word prediction.

In this study, we propose the Ductal Instance-Oriented Pipeline (DIOP) that contains a duct-level instance segmentation model, a tissue-level semantic segmentation model, and three-levels of features for diagnostic classification. Based on recent advancements in instance segmentation and the Mask R-CNN model, our duct-level segmenter tries to identify each ductal individual inside a microscopic image; then, it extracts tissue-level information from the identified ductal instances. Leveraging three levels of information obtained from these ductal instances and also the histopathology image, the proposed DIOP outperforms previous approaches (both feature-based and CNN-based) in all diagnostic tasks; for the four-way classification task, the DIOP achieves comparable performance to general pathologists in this unique dataset. The proposed DIOP only takes a few seconds to run in the inference time, which could be used interactively on most modern computers. More clinical explorations are needed to study the robustness and generalizability of this system in the future., Comment: ICPR 2020. Submitted July 15th, 2020

High resolution histology images contain information related to disease prognosis. However, survival prediction based on current clinical grading systems, which rely heavily on a pathologist’s histological assessment, has significant limitations due to the heterogeneity and complexity of tissue phenotypes. To address these challenges, we propose a deep learning framework that leverages hierarchical graph-based representations to enable more precise prediction of progression-free survival in prostate cancer patients. Unlike conventional approaches that analyze patch-based or cell-based pathomic features alone without considering their spatial connectivity, we explore multi-scale topological structures of whole slide images in an integrative context. Extensive experiments have demonstrated the effectiveness of our model for better progression prediction.

The Dirichlet and hierarchical Dirichlet processes are two important techniques for nonparametric Bayesian learning. These learning techniques allow unsupervised learning without specifying traditionally used input parameters. In topic modeling, this can be applied to discovering topics without specifying the number beforehand. Existing methods, such as those applied to topic modeling, usually take on a complex sampling calculation for inference. These techniques for inference of the Dirichlet and hierarchal Dirichlet processes are often based on Markov processes that can deviate from parametric topic modeling. This deviation may not be the best approach in the context of nonparametric topic modeling. Additionally, since they often rely on approximations they can negatively affect the predictive power of such models. In this paper we introduce a new interpretation of nonparametric Bayesian learning called the biased coin flip process—contrived for use in the context of Bayesian topic modeling. We prove mathematically the equivalence of the biased coin flip process to the Dirichlet process with an additional parameter representing the number of trials. A major benefit of the biased coin flip process is the similarity of the inference calculation to that of previous established parametric topic models—which we hope will lead to a more widespread adoption of hierarchical Dirichlet process based topic modeling. Additionally, as we show empirically the biased coin flip process leads to a nonparametric topic model with improved predictive performance.

Purpose: Prostate cancer (PCa) is the most common solid organ cancer and second leading cause of death in men. Multiparametric magnetic resonance imaging (mpMRI) enables detection of the most aggressive, clinically significant PCa (csPCa) tumors that require further treatment. A suspicious region of interest (ROI) detected on mpMRI is now assigned a Prostate Imaging-Reporting and Data System (PIRADS) score to standardize interpretation of mpMRI for PCa detection. However, there is significant inter-reader variability among radiologists in PIRADS score assignment and a minimal input semi-automated artificial intelligence (AI) system is proposed to harmonize PIRADS scores with mpMRI data. Approach: The proposed deep learning model (the seed point model) uses a simulated single-click seed point as input to annotate the lesion on mpMRI. This approach is in contrast to typical medical AI-based approaches that require annotation of the complete lesion. The mpMRI data from 617 patients used in this study were prospectively collected at a major tertiary U.S. medical center. The model was trained and validated to classify whether an mpMRI image had a lesion with a PIRADS score greater than or equal to PIRADS 4. Results: The model yielded an average receiver-operator characteristic (ROC) area under the curve (ROC-AUC) of 0.704 over a 10-fold cross-validation, which is significantly higher than the previously published benchmark. Conclusions: The proposed model could aid in PIRADS scoring of mpMRI, providing second reads to promote quality as well as offering expertise in environments that lack a radiologist with training in prostate mpMRI interpretation. The model could help identify tumors with a higher PIRADS for better clinical management and treatment of PCa patients at an early stage.

Deep learning methods have exhibited impressive performance in many classification and prediction tasks in healthcare using electronic health record (EHR) data. However, effectively handling the heterogeneous nature and sparsity of EHR data remain as challenges. In this work, we present a model that utilizes heterogeneous data and attempts to address sparsity using temporal Multi-Level Embeddings of diagnoses, procedures, and medication codes with demographic information and Topic modeling (MLET). MLET aggregates various categories of EHR data and learns inherent structure based on hospital visits in a patient’s medical history. We demonstrate the potential of the approach in the task of predicting depression using different time windows prior to a clinical diagnosis. We found that MLET outperformed all baseline methods with a highest improvement from 0.6122 to 0.6808 in precision recall area under the curve (PRAUC). Our results demonstrate the model’s ability to utilize heterogeneous EHR information to predict depression, which may have future implications for screening and early detection.

Background Patients with stable ischemic heart disease represent a heterogeneous population at variable risk for major adverse cardiac events (MACE). Because MACE typically occurs outside the hospital, we studied whether biometric and psychometric remote patient monitoring are associated with MACE risk biomarkers. Methods and Results In 198 patients with stable ischemic heart disease (mean age 65±11 years, 60% women), we evaluated baseline measures, including biometric (FitBit 2) and psychometric (acquired via smartphone‐administered patient‐reported outcomes) remote monitoring, in the PRE‐MACE (Prediction, Risk, and Evaluation of Major Adverse Cardiac Events) study. In multivariable adjusted regression analyses, we examined the association of these measures with biomarkers of MACE risk, including NT‐proBNP (N‐terminal pro‐b‐type natriuretic peptide), u‐hs‐cTnI (ultra‐high sensitivity cardiac‐specific troponin I), and hs‐CRP (high‐sensitivity C‐reactive) protein. Both biometric and psychometric measures were associated with NT‐proBNP. Specifically, step count, heart rate, physical activity, global health score, and physical function score were all inversely related, whereas physical limitation score was directly related ( P ≤0.05 for all). However, only biometric measures (step count and heart rate) were associated with u‐hs‐cTnI (inversely related, P P ≤0.05). Conclusions In stable ischemic heart disease patients, remotely monitored measures were associated with MACE risk biomarkers. Both biometric and psychometric measures were related to NT‐proBNP. In contrast, biometric measures were uniquely related to u‐hs‐cTnI, while psychometric indices were uniquely related to hs‐CRP. Further investigation could assess the predictive value of these metrics for MACE in ischemic heart disease.

In this work, we investigate semi-supervised learning (SSL) for image classification using adversarial training. Previous results have illustrated that generative adversarial networks (GANs) can be used for multiple purposes. Triple-GAN, which aims to jointly optimize model components by incorporating three players, generates suitable image-label pairs to compensate for the lack of labeled data in SSL with improved benchmark performance. Conversely, Bad (or complementary) GAN, optimizes generation to produce complementary data-label pairs and force a classifier's decision boundary to lie between data manifolds. Although it generally outperforms Triple-GAN, Bad GAN is highly sensitive to the amount of labeled data used for training. Unifying these two approaches, we present unified-GAN (UGAN), a novel framework that enables a classifier to simultaneously learn from both good and bad samples through adversarial training. We perform extensive experiments on various datasets and demonstrate that UGAN: 1) achieves state-of-the-art performance among other deep generative models, and 2) is robust to variations in the amount of labeled data used for training.

Bone age assessment (BAA) is clinically important as it can be used to diagnose endocrine and metabolic disorders during child development. Existing deep learning based methods for classifying bone age use the global image as input, or exploit local information by annotating extra bounding boxes or key points. However, training with the global image underutilizes discriminative local information, while providing extra annotations is expensive and subjective. In this paper, we propose an attention-guided approach to automatically localize the discriminative regions for BAA without any extra annotations. Specifically, we first train a classification model to learn the attention maps of the discriminative regions, finding the hand region, the most discriminative region (the carpal bones), and the next most discriminative region (the metacarpal bones). Guided by those attention maps, we then crop the informative local regions from the original image and aggregate different regions for BAA. Instead of taking BAA as a general regression task, which is suboptimal due to the label ambiguity problem in the age label space, we propose using joint age distribution learning and expectation regression, which makes use of the ordinal relationship among hand images with different individual ages and leads to more robust age estimation. Extensive experiments are conducted on the RSNA pediatric bone age data set. Using no training annotations, our method achieves competitive results compared with existing state-of-the-art semi-automatic deep learning-based methods that require manual annotation. Code is available at https: //github.com/chenchao666/Bone-Age-Assessment., codes are available at https://github.com/chenchao666/Bone-Age-Assessment

Objective To evaluate the level of evidence behind recommendations on social media for disease prevention in five lower urinary tract symptoms. Materials and Methods We conducted a digital analysis of anonymous online posts on social media sites collected by a social media data mining service. One thousand posts about pelvic organ prolapse, stress urinary incontinence, overactive bladder, urinary tract infection, and interstitial cystitis/bladder pain syndrome were randomly selected. We analyzed these posts for recommendations regarding the prevention and treatment of these diseases, which were then compared to recommendations in available clinical guidelines and assessed for level of evidence. Results A total of 158 of 1000 posts contained 239 prevention strategies. For pelvic organ prolapse, there were 41 strategies identified, 25 (61%) of which had no evidence. For urinary tract infection 14 of 58 (29%) had no evidence, including recommendations for dietary modifications and urinary alkalization. For overactive bladder 8 of 28 (29%) had level 4 or no evidence. For stress urinary incontinence, 12 of 34 (36%) of prevention strategies had no evidence, such as laser rejuvenation and bladder training. Interstitial cystitis had the highest number of prevention strategies, and most were low or nonevidence based (70/79, 89%). Conclusion Prevention and treatment strategies are common in online discussions of pelvic floor disorders, but at least one third of these recommendations have no evidential support. There is a role for further online education and social media engagement by health care specialists to promote evidence-based practices.

BackgroundHeart failure is a debilitating disease affecting millions of adults in the United States, often leading to hospitalizations and hospital readmissions. Around 50% of readmissions due to heart failure are preventable, with lack of adherence to prescribed self-care as a driving factor. Remote tracking and reminders issued by mobile health devices such as activity trackers and smartphone apps could help to promote self-care, which could potentially reduce these readmissions.ObjectiveThis pilot study used minimally-invasive monitoring technologies and patient-reported outcomes to examine the pragmatic feasibility of a remote monitoring regimen.MethodsTwenty heart failure patients participated in piloting a remote monitoring regimen. Data collection included: (1) continuous remote monitoring using wrist-worn consumer activity trackers; (2) body weight recording using bathroom scales; (3) dose tracking using smart pill bottles; and (4) patient-reported outcome measures.ResultsParticipants were aged 65.3 years on average, 50.0% of the participants were women, and 81.8% of the participants were determined to be New York Heart Association Class III or higher. Over the course of the study, 60.0% of the subjects wore the activity tracker for at least 70% of the hours, and 45.0% used the bathroom scale for more than 70% of the days. For the smart pill bottle, 55.0% of the subjects used it less than 10% of the days. Usage of the activity tracker correlated significantly with changes in Self-Care of Heart Failure Index confidence subscale scores and changes in Seattle Angina Questionnaire scores (P< .05), and usage of the bathroom scale correlated significantly with changes in Seattle Angina Questionnaire scores (P= .04).ConclusionsThe majority of the participants maintained a high adherence to wearing the activity tracker, but had low adherence to using the smart pill bottle. Usage of the bathroom scale was fair, but it received positive reviews from most subjects. Given the observed usage and feedback, we suggest that mobile health-driven interventions consider including an activity tracker and bathroom scale. Furthermore, the data’s correlations with changes in patient-reported outcomes indicate the potential for these devices to be an effective way to remotely monitor heart failure patients.

Treatment of acute ischemic strokes (AIS) is largely contingent upon the time since stroke onset (TSS). However, TSS may not be readily available in up to 25% of patients with unwitnessed AIS. Current clinical guidelines for patients with unknown TSS recommend the use of MRI to determine eligibility for thrombolysis, but radiology assessments have high inter-reader variability. In this work, we present deep learning models that leverage MRI diffusion series to classify TSS based on clinically validated thresholds. We propose an intra-domain task-adaptive transfer learning method, which involves training a model on an easier clinical task (stroke detection) and then refining the model with different binary thresholds of TSS. We apply this approach to both 2D and 3D CNN architectures with our top model achieving an ROC-AUC value of 0.74, with a sensitivity of 0.70 and a specificity of 0.81 for classifying TSS < 4.5 hours. Our pretrained models achieve better classification metrics than the models trained from scratch, and these metrics exceed those of previously published models applied to our dataset. Furthermore, our pipeline accommodates a more inclusive patient cohort than previous work, as we did not exclude imaging studies based on clinical, demographic, or image processing criteria. When applied to this broad spectrum of patients, our deep learning model achieves an overall accuracy of 75.78% when classifying TSS < 4.5 hours, carrying potential therapeutic implications for patients with unknown TSS.

Advancements in machine learning algorithms have had a beneficial impact on representation learning, classification, and prediction models built using electronic health record (EHR) data. Effort has been put both on increasing models' overall performance as well as improving their interpretability, particularly regarding the decision-making process. In this study, we present a temporal deep learning model to perform bidirectional representation learning on EHR sequences with a transformer architecture to predict future diagnosis of depression. This model is able to aggregate five heterogenous and high-dimensional data sources from the EHR and process them in a temporal manner for chronic disease prediction at various prediction windows. We applied the current trend of pretraining and fine-tuning on EHR data to outperform the current state-of-the-art in chronic disease prediction, and to demonstrate the underlying relation between EHR codes in the sequence. The model generated the highest increases of precision-recall area under the curve (PRAUC) from 0.70 to 0.76 in depression prediction compared to the best baseline model. Furthermore, the self-attention weights in each sequence quantitatively demonstrated the inner relationship between various codes, which improved the model's interpretability. These results demonstrate the model's ability to utilize heterogeneous EHR data to predict depression while achieving high accuracy and interpretability, which may facilitate constructing clinical decision support systems in the future for chronic disease screening and early detection., Comment: in IEEE Journal of Biomedical and Health Informatics (2021)

Semantic segmentation of histopathology images can be a vital aspect of computer-aided diagnosis, and deep learning models have been effectively applied to this task with varying levels of success. However, their impact has been limited due to the small size of fully annotated datasets. Data augmentation is one avenue to address this limitation. Generative Adversarial Networks (GANs) have shown promise in this respect, but previous work has focused mostly on classification tasks applied to MR and CT images, both of which have lower resolution and scale than histopathology images. There is limited research that applies GANs as a data augmentation approach for large-scale image semantic segmentation, which requires high-quality image-mask pairs. In this work, we propose a multi-scale conditional GAN for high-resolution, large-scale histopathology image generation and segmentation. Our model consists of a pyramid of GAN structures, each responsible for generating and segmenting images at a different scale. Using semantic masks, the generative component of our model is able to synthesize histopathology images that are visually realistic. We demonstrate that these synthesized images along with their masks can be used to boost segmentation performance, especially in the semi-supervised scenario.

We conducted a large-scale digital ethnographic analysis of anonymous online posts to capture the complete patient experience, knowledge and perceptions among women with stress urinary incontinence.Online posts were identified through data mining. Overall, 200 randomized posts were analyzed using grounded theory qualitative methods. To ensure full thematic discovery we also applied a Latent Dirichlet Allocation probabilistic topic modeling approach to the entire data set of identified posts. Latent Dirichlet Allocation topics are represented as a distribution of words, similar to a word cloud, which were manually reviewed to identify themes.A total of 985 online posts by 762 unique users were extracted from 98 websites. There was significant overlap between the grounded theory and Latent Dirichlet Allocation identified themes. Our analysis suggests that these online communities help women manage the quality of life impact of their stress urinary incontinence, navigate specialty care and reach a decision regarding surgical vs nonsurgical management. Additionally, we identified risk factors, prevention strategies and treatment recommendations discussed online.Findings demonstrated patient values that may influence decision making when seeking care for stress urinary incontinence and choosing a treatment. Social media interactions provide insight into patient behaviors that are important in order to improve patient centered care and decision making.

INTRODUCTION AND OBJECTIVE:Qualitative methods assessing women’s perspectives on living with OAB have traditionally been obtained via focus groups and interviews. Now a plethora of cyber forums hav...

Background Heart failure is a serious public health concern that afflicts millions of individuals in the United States. Development of behaviors that promote heart failure self-care may be imperative to reduce complications and avoid hospital re-admissions. Mobile health solutions, such as activity trackers and smartphone apps, could potentially help to promote self-care through remote tracking and issuing reminders. Objective The objective of this study was to ascertain heart failure patients’ interest in a smartphone app to assist them in managing their treatment and symptoms and to determine factors that influence their interest in such an app. Methods In the clinic waiting room on the day of their outpatient clinic appointments, 50 heart failure patients participated in a self-administered survey. The survey comprised 139 questions from previously published, institutional review board–approved questionnaires. The survey measured patients’ interest in and experience using technology as well as their function, heart failure symptoms, and heart failure self-care behaviors. The Minnesota Living with Heart Failure Questionnaire (MLHFQ) was among the 11 questionnaires and was used to measure the heart failure patients’ health-related quality of life through patient-reported outcomes. Results Participants were aged 64.5 years on average, 32% (16/50) of the participants were women, and 91% (41/45) of the participants were determined to be New York Heart Association Class II or higher. More than 60% (30/50) of the survey participants expressed interest in several potential features of a smartphone app designed for heart failure patients. Participant age correlated negatively with interest in tracking, tips, and reminders in multivariate regression analysis (P Conclusions The majority of heart failure patients showed interest in activity tracking, heart failure symptom management tips, and reminder features of a smartphone app. Desirable features and an understanding of factors that influence patient interest in a smartphone app for heart failure self-care may allow researchers to address common concerns and to develop apps that demonstrate the potential benefits of mobile technology.

Constructing statistical models using personal sensor data could allow for tracking health status over time, thereby enabling the possibility of early intervention. The goal of this study was to use machine learning algorithms to classify patient-reported outcomes (PROs) using activity tracker data in a cohort of patients with stable ischemic heart disease (SIHD). A population of 182 patients with SIHD were monitored over a period of 12 weeks. Each subject received a Fitbit Charge 2 device to record daily activity data, and each subject completed eight Patient-Reported Outcomes Measurement Information Systems (PROMIS®) short form at the end of each week as a self-assessment of their health status. Two models were built to classify PRO scores using activity tracker data. The first model treated each week independently, while the second used a Hidden Markov model (HMM) to take advantage of correlations between successive weeks. Retrospective analysis compared the classification accuracy of the two models and the importance of each feature. In the independent model, a random forest classifier achieved a mean area under curve (AUC) of 0.76 for classifying the Physical Function PRO. The HMM model achieved significantly better AUCs for all PROs (p

INTRODUCTION AND OBJECTIVES:Focal therapies such as hemi-gland ablation (HGA) are evolving treatment options for prostate cancer. However, limitations in current diagnostics--biopsy Gleason score, ...

INTRODUCTION AND OBJECTIVES:Although pelvic floor disorders are common, women do not feel comfortable discussing their condition with others, including providers. Many turn to online support groups...

INTRODUCTION AND OBJECTIVES:Characterizing women's perceptions of urinary tract infections (UTI) helps patients achieve a better understanding of their disease and increases the likelihood of treat...

INTRODUCTION AND OBJECTIVES:Women′s experience with overactive bladder (OAB) has been previously characterized with focus groups and online surveys designed from the perspective of medical provider...

INTRODUCTION AND OBJECTIVES:Most qualitative studies of women stress urinary incontinence (SUI) have relied on in-person interviews and surveys, some of which have been limited by inclusion of one ...

INTRODUCTION AND OBJECTIVES:Although pelvic organ prolapse (POP) has a prevalence of 3-8%, feelings of shame and confusion have kept women from addressing their concerns with medical providers. Pre...

INTRODUCTION AND OBJECTIVES:MRI-targeted biopsy has improved the accuracy of prostate cancer diagnosis, but regions of interest (ROIs) are often labeled as equivocal category 3 lesions based on PIR...

Current clinical practice relies on clinical history to determine the time since stroke (TSS) onset. Imaging-based determination of acute stroke onset time could provide critical information to clinicians in deciding stroke treatment options, such as thrombolysis. The patients with unknown or unwitnessed TSS are usually excluded from thrombolysis, even if their symptoms began within the therapeutic window. In this paper, we demonstrate a machine learning approach for TSS classification using routinely acquired imaging sequences. We develop imaging features from the magnetic resonance (MR) images and train machine learning models to classify the TSS. We also propose a deep-learning model to extract hidden representations for the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional deep features. The cross-validation results show that our best classifier achieved an area under the curve of 0.765, with a sensitivity of 0.788 and a negative predictive value of 0.609, outperforming existing methods. We show that the features generated by our deep-learning algorithm correlate with the MR imaging features, and validate the robustness of the model on imaging parameter variations (e.g., year of imaging). This paper advances magnetic resonance imaging analysis one-step-closer to an operational decision support tool for stroke treatment guidance.

Objective Few studies have examined ankylosing spondylitis (AS) patients' concerns about and perceptions of biologic therapies, apart from traditional surveys. In this study, we used social media data to examine the knowledge, attitudes, and beliefs of AS patients regarding biologic therapies. Methods We collected posts published on 601 social media sites between January 1, 2016 and April 26, 2017. In each post, both an AS keyword and a biologic were mentioned. To explore themes within the collection of posts in an unsupervised manner, a latent Dirichlet allocation topic model was fit to the data set. Each discovered topic was represented as a discrete distribution over the words in the collection, similar to a word cloud. The topics were manually reviewed to identify themes, which were confirmed using thematic data analysis. Results We examined 27,416 social media posts and identified 112 themes. The majority of themes (n = 67 [60%]) focused on discussions related to AS treatment. Other themes, including the psychological impact of AS, reporting of medical literature, and AS disease consequences, accounted for the remaining 40% (n = 45). In discussions regarding AS treatment, most topics involved biologics, and most subthemes involved side effects (e.g., fatigue, allergic reactions), biologic treatment attributes (e.g., dosing, frequency), and concerns about use of biologics (e.g., increased cancer risk). Additional implicit patient needs (e.g., support) were identified using qualitative analyses. Conclusion Social media revealed a dynamic range of themes governing AS patients' experience with and choice of biologic agents. The complexity of selecting biologics from among many such agents and navigating their risk/benefit profiles suggests the merit of creating online tools tailored to support patients' decision-making with regard to biologic therapies for AS.

Hemorrhagic transformation (HT) is one of the most devastating complications of reperfusion therapy in acute ischemic stroke. Prediction of an upcoming HT remains beyond current techniques in routine clinical practice. If made available, such information would benefit the management of acute ischemic stroke patients and help to tailor therapeutic strategies. This study aims at providing a machine learning framework for predicting occurrence and extent of HT from source perfusion-weighted magnetic resonance imaging (PWI) combined with diffusion weighted imaging (DWI). The model relies on a LSTM network based on PWI combined with DWI imaging features into a fully connected neural network. A retrospective comparative analysis performed on 155 acute stroke patients demonstrate the efficacy of the LSTM model (AUC-ROC: \(89.4\%\)) against state-of-the-art machine learning models. Predicted likelihood of HT at the voxel level was evaluated against HT annotations of stroke neurologists obtained from follow-up gradient recalled echo (GRE) imaging.

We describe the protocol, design, and methodology of the Prediction, Risk, and Evaluation of Major Adverse Cardiac Events (PRE-MACE) study as a multicomponent remote patient monitoring in cardiology. Using biosensor, biomarkers, and patient-reported outcomes in participants with stable ischemic heart disease, the PRE-MACE study is designed to measure cross-sectional correlations and establish the ability of remote monitoring to predict major adverse cardiovascular event (MACE) biomarkers and incident MACE at baseline and 12-month follow-up. It will further assess the adherence and cost-effectiveness of remote monitoring and blood sampling over the initial months. Despite medication and lifestyle changes, patients with cardiovascular disease can experience MACE due to undertreatment, poor adherence, or failure to recognize clinical or biochemical changes that presage MACE. Identifying patients using remote monitoring to detect MACE forerunners has potential to improve outcomes, avoid MACE, and reduce resource utilization. Data collection will include: (1) continuous remote monitoring using wearable biosensors; (2) biomarker measurements using plasma and at-home micro-sampling blood collection; and (3) patient-reported outcomes to monitor perceived stress, anxiety, depression, and health-related quality of life. Two hundred participants will be followed for 90 days with a subset (n = 80) monitored for 180 days. All participants will be followed up for MACE at 12 months.The PRE-MACE study will utilize remote monitoring with biosensors, biomarkers, and patient-reported outcomes to identify intermediate biomarkers of MACE in patients with stable ischemic heart disease. If shown to be effective, this intervention can be utilized between health visits to predict MACE and reduce financial impact of MACE.

Rationale and Objectives: The current paradigm of cancer diagnosis involves uncoordinated communication of findings from radiology and pathology to downstream physicians. Discordance between these findings can require additional time from downstream users to resolve, or given incorrect resolution, may adversely impact treatment decisions. To mitigate this problem, we developed a webbased system, called RadPath, for correlating and integrating radiology and pathology reporting. Materials and Methods: RadPath includes interfaces to our institution’s clinical information systems, which are used to retrieve reports, images, and test results that are structured into an interactive compendium for a diagnostic patient case. The system includes an editing interface for physicians, allowing for the inclusion of additional clinical data, as well as the ability to retrospectively correlate and contextualize imaging findings following pathology diagnosis. Results: During pilot deployment and testing over the course of 1 year, physicians at our institution have completed 60 RadPath cases, requiring an average of 128 seconds from a radiologist and an average of 93 seconds from a pathologist per case. Several technical and workflow challenges were encountered during development, including interfacing with diverse clinical information systems, automatically structuring report contents, and determining the appropriate physicians to create RadPath summaries. Reaction to RadPath has been positive, with users valuing the system’s ability to consolidate diagnostic information. Conclusions: With the increasing complexity of medicine and the movement toward team-based disease management, there is a need for improved clinical communication and information exchange. RadPath provides a platform for generating coherent and correlated diagnostic summaries in cancer diagnosis with minimal additional effort from physicians.

Prostate cancer is the most common and second most deadly form of cancer in men in the United States. The classification of prostate cancers based on Gleason grading using histological images is important in risk assessment and treatment planning for patients. Here, we demonstrate a new region-based convolutional neural network framework for multi-task prediction using an epithelial network head and a grading network head. Compared with a single-task model, our multi-task model can provide complementary contextual information, which contributes to better performance. Our model is achieved a state-of-the-art performance in epithelial cells detection and Gleason grading tasks simultaneously. Using fivefold cross-validation, our model is achieved an epithelial cells detection accuracy of 99.07% with an average area under the curve of 0.998. As for Gleason grading, our model is obtained a mean intersection over union of 79.56% and an overall pixel accuracy of 89.40%.

Telemedicine has been used to remotely diagnose and treat patients, yet previously applied telemonitoring approaches have been fraught with adherence issues. The primary goal of this study was to evaluate the adherence rates using a consumer-grade continuous-time heart rate and activity tracker in a mid-risk cardiovascular patient population. As a secondary analysis, we show the ability to utilize the information provided by this device to identify information about a patient’s state by correlating tracker information with patient-reported outcome survey scores. We showed that using continuous-time activity trackers with heart rate monitors can be effective in a telemonitoring application, as patients had a high level of adherence (90.0% median usage) and low attrition (0.09% decrease per day) over a 90-day period. Furthermore, data collected correlated significantly with clinically relevant patient surveys (r2=0.15 for PROMIS global health scores, p

Certain pathology workflows, such as classification and grading of prostate adenocarcinoma according to the Gleason grade scheme, stand to gain speed and objectivity by incorporating contemporary digital image analysis methods. We compiled a dataset of 513 high resolution image tiles from primary prostate adenocarcinoma wherein individual glands and stroma were demarcated and graded by hand. With this unique dataset, we tested four Convolutional Neural Network architectures including FCN-8s, two SegNet variants, and multi-scale U-Net for performance in semantic segmentation of high- and low-grade tumors. In a 5-fold cross-validation experiment, the FCN-8s architecture achieved a mIOU of 0.759 and an accuracy of 0.87, while the less complex U-Net architecture achieved a mIOU of 0.738 and accuracy of 0.885. The FCN-8s architecture applied to whole slide images not used for training achieved a mIOU of 0.857 in annotated tumor foci with a multiresolution processing time averaging 11 minutes per slide. The three architectures tested on whole slides all achieved areas under the Receiver Operating Characteristic curve near 1, strongly demonstrating the suitability of semantic segmentation Convolutional Neural Networks for detecting and grading prostate cancer foci in radical prostatectomies.

Automated Gleason grading is an important preliminary step for quantitative histopathological feature extraction. Different from the traditional task of classifying small pre-selected homogeneous regions, semantic segmentation provides pixel-wise Gleason predictions across an entire slide. Deep learning-based segmentation models can automatically learn visual semantics from data, which alleviates the need for feature engineering. However, performance of deep learning models is limited by the scarcity of large-scale fully annotated datasets, which can be both expensive and time-consuming to create. One way to address this problem is to leverage external weakly labeled datasets to augment models trained on the limited data. In this paper, we developed an expectation maximization-based approach constrained by an approximated prior distribution in order to extract useful representations from a large number of weakly labeled images generated from low-magnification annotations. This method was utilized to improve the performance of a model trained on a limited fully annotated dataset. Our semi-supervised approach trained with 135 fully annotated and 1,800 weakly annotated tiles achieved a mean Jaccard Index of 49.5% on an independent test set, which was 14% higher than the initial model trained only on the fully annotated dataset.

The P300 Speller Brain-Computer Interface (BCI) provides a means of communication for those suffering from advanced neuromuscular diseases such as amyotrophic lateral sclerosis (ALS). Recent literature has incorporated language-based modelling, which uses previously chosen characters and the structure of natural language to modify the interface and classifier. Two complementary methods of incorporating language models have previously been independently studied: predictive spelling uses language models to generate suggestions of complete words to allow for the selection of multiple characters simultaneously, and language model-based classifiers have used prior characters to create a prior probability distribution over the characters based on how likely they are to follow. In this study, we propose a combined method which extends a language-based classifier to generate prior probabilities for both individual characters and complete words. In order to gauge the efficiency of this new model, results across 12 healthy subjects were measured. Incorporating predictive spelling increased typing speed using the P300 speller, with an average increase of 15.5% in typing rate across subjects, demonstrating that language models can be effectively utilized to create full word suggestions for predictive spelling. When combining predictive spelling with language model classification, typing speed is significantly improved, resulting in better typing performance.