Author name cluster

William Speier

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

1 author row

JBHI Journal 2022 Journal Article

Attention-Guided Discriminative Region Localization and Label Distribution Learning for Bone Age Assessment

Chao Chen
Zhihong Chen
Xinyu Jin
Lanjuan Li
William Speier
Corey W. Arnold

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. Without using extra manual annotations, our method achieves competitive results compared with existing state-of-the-art deep learning-based methods that require manual annotation. Code is available at https://github.com/chenchao666/Bone-Age-Assessment.

Details DOI

JBHI Journal 2021 Journal Article

Bidirectional Representation Learning From Transformers Using Multimodal Electronic Health Record Data to Predict Depression

Yiwen Meng
William Speier
Michael K. Ong
Corey W. Arnold

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.

Details DOI

JBHI Journal 2021 Journal Article

HCET: Hierarchical Clinical Embedding With Topic Modeling on Electronic Health Records for Predicting Future Depression

Yiwen Meng
William Speier
Michael Ong
Corey W. Arnold

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.

Details DOI

JBHI Journal 2020 Journal Article

A Machine Learning Approach to Classifying Self-Reported Health Status in a Cohort of Patients With Heart Disease Using Activity Tracker Data

Yiwen Meng
William Speier
Chrisandra Shufelt
Sandy Joung
Jennifer E Van Eyk
C. Noel Bairey Merz
Mayra Lopez
Brennan Spiegel

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 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, whereas 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 < 0. 05) other than Fatigue and Sleep Disturbance, with a highest mean AUC of 0. 79 for the physical function-short form 10a. Our study demonstrates the ability of activity tracker data to classify health status over time. These results suggest that patient outcomes can be monitored in real time using activity trackers.

Details DOI

YNICL Journal 2018 Journal Article

The potential value of probabilistic tractography-based for MR-guided focused ultrasound thalamotomy for essential tremor

Evangelia Tsolaki
Angela Downes
William Speier
W. Jeff Elias
Nader Pouratian

Magnetic Resonance-guided Focused UltraSound (MRgFUS) offers an incisionless approach to treat essential tremor (ET). Due to lack of evident internal anatomy on traditional structural imaging, indirect targeting must still be used to localize the lesion. Here, we investigate the potential predictive value of probabilistic tractography guided thalamic targeting by defining how tractography-defined targets, lesion size and location, and clinical outcomes interrelate. MR imaging and clinical outcomes from 12 ET patients that underwent MRgFUS thalamotomy in a pilot study at the University of Virginia were evaluated in this analysis. FSL was used to evaluate each patient's voxel-wise thalamic connectivity with FreeSurfer generated pre- and post-central gyrus targets, to generate thalamic target maps. Using Receiver Operating Characteristic curves, the overlap between these thalamic target maps and the MRgFUS lesion was systematically evaluated relative to clinical outcome. To further define the connectivity characteristics of effective MRgFUS thalamotomy lesions, we evaluated whole brain probabilistic tractography of lesions (using post-treatment imaging to define the lesion pre-treatment diffusion tensor MRI). The structural connectivity difference was explored between subjects with the best clinical outcome relative to all others. Ten of twelve patients presented high percentage of overlapping between connectivity-based thalamic segmentation maps and lesion area. The improvement of clinical score was predicted (AUC: 0.80) using the volume of intersection between the thalamic target (precentral gyrus) map and MRgFUS induced lesion as feature. The main structural differences between those with different magnitudes of response were observed in connectivity to the pre- and post-central gyri and brainstem/cerebellum. MRgFUS thalamotomy lesions characterized by strong structural connectivity to precentral gyrus demonstrated better responses in a cohort of patients treated with MRgFUS for ET. The intersection between lesion and thalamic-connectivity maps to motor - sensory targets proved to be effective in predicting the response to the therapy. These imaging techniques can be used to increase the efficacy and consistency of outcomes with MRgFUS and potentially shorten treatment times by identifying optimal targets in advance of treatment.

Details DOI