Joseph Y Lo

Duke University Medical Center | DUMC · Department of Radiology

Objectives To assess the performance bias caused by sampling data into training and test sets in a mammography radiomics study. Methods Mammograms from 700 women were used to study upstaging of ductal carcinoma in situ. The dataset was repeatedly shuffled and split into training (n = 400) and test cases (n = 300) forty times. For each split, cross...

Ductal carcinoma in situ (DCIS) is a very common non-life threatening, pre-invasive form of breast cancer constituting 25% of all new breast cancer diagnoses in the USA, and is normally treated with invasive measures. However, only 20 to 30% of DCIS cases will progress to life-threatening invasive breast cancer in their lifetime. There is a need fo...

Many studies have investigated deep-learning-based artificial intelligence (AI) models for medical imaging diagnosis of the novel coronavirus (COVID-19), with many reports of near-perfect performance. However, variability in performance and underlying data biases raise concerns about clinical generalizability. This retrospective study involved the...

Computer-aided detection (CAD) frameworks for breast cancer screening have been researched for several decades. Early adoption of deep-learning models in CAD frameworks has shown greatly improved detection performance compared to traditional CAD on single-view images. Recently, studies have improved performance by merging information from multiple...

Rationale and Objectives Adoption of the Prostate Imaging Reporting & Data System (PI-RADS) has been shown to increase detection of clinically significant prostate cancer on prostate mpMRI. We propose that a rule-based algorithm based on Regular Expression (RegEx) matching can be used to automatically categorize prostate mpMRI reports into categori...

Research studies of artificial intelligence models in medical imaging have been hampered by poor generalization. This problem has been especially concerning over the last year with numerous applications of deep learning for COVID-19 diagnosis. Virtual imaging trials (VITs) could provide a solution for objective evaluation of these models. In this w...

Organ segmentation of medical images is a key step in virtual imaging trials. However, organ segmentation datasets are limited in terms of quality (because labels cover only a few organs) and quantity (since case numbers are limited). In this study, we explored the tradeoffs between quality and quantity. Our goal is to create a unified approach for...

Despite the potential of weakly supervised learning to automatically annotate massive amounts of data, little is known about its limitations for use in computer-aided diagnosis (CAD). For CT specifically, interpreting the performance of CAD algorithms can be challenging given the large number of co-occurring diseases. This paper examines the effect...

Purpose: The purpose of this work was to characterize and improve the ability of fused filament fabrication to create anthropomorphic physical phantoms for CT research. Specifically, we sought to develop the ability to create multiple levels of x-ray attenuation with a single material. Methods: CT images of 3D printed cylinders with different in...

In [1], the dose estimation accuracy using the alternative baseline method under modulated tube current was not correctly calculated due to an unintentional simulation error.

Purpose: To design multi-disease classifiers for body CT scans for three different organ systems using automatically extracted labels from radiology text reports. Materials & Methods: This retrospective study included a total of 12,092 patients (mean age 57 ± 18; 6,172 women) for model development and testing (from 2012-2017). Rule-based algorith...

Interpretability in machine learning models is important in high-stakes decisions such as whether to order a biopsy based on a mammographic exam. Mammography poses important challenges that are not present in other computer vision tasks: datasets are small, confounding information is present and it can be difficult even for a radiologist to decide...

In mammography and tomosynthesis, radiologists use the geometric relationship of the four standard screening views to detect breast abnormalities. To date, computer aided detection methods focus on formulations based only on a single view. Recent multi-view methods are either black box approaches using methods such as relation blocks, or perform ex...

Importance Breast cancer screening is among the most common radiological tasks, with more than 39 million examinations performed each year. While it has been among the most studied medical imaging applications of artificial intelligence, the development and evaluation of algorithms are hindered by the lack of well-annotated, large-scale publicly av...

Background: The strategy to combat the problem associated with large deformations in the breast due to the difference in the medical imaging of patient posture plays a vital role in multimodal medical image registration with artificial intelligence (AI) initiatives. How to build a breast biomechanical model simulating the large-scale deformation o...

When we deploy machine learning models in high-stakes medical settings, we must ensure these models make accurate predictions that are consistent with known medical science. Inherently interpretable networks address this need by explaining the rationale behind each decision while maintaining equal or higher accuracy compared to black-box models. In...

Most tissue collections of neoplasms are composed of formalin-fixed and paraffin-embedded (FFPE) excised tumor samples used for routine diagnostics. DNA sequencing is becoming increasingly important in cancer research and clinical management; however it is difficult to accurately sequence DNA from FFPE samples. We developed and validated a new bioi...

Interpretability in machine learning models is important in high-stakes decisions, such as whether to order a biopsy based on a mammographic exam. Mammography poses important challenges that are not present in other computer vision tasks: datasets are small, confounding information is present, and it can be difficult even for a radiologist to decid...

To develop a high throughput multi-label annotator for body Computed Tomography (CT) reports that can be applied to a variety of diseases, organs, and cases. First, we used a dictionary approach to develop a rule-based algorithm (RBA) for extraction of disease labels from radiology text reports. We targeted three organ systems (lungs/pleura, liver/...

Machine learning models for radiology benefit from large-scale data sets with high quality labels for abnormalities. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 19,993 unique patients. This is the largest multiply-annotated volumetric medical imaging data set reported. To annotate this data set, we devel...

Breast cancer screening is one of the most common radiological tasks with over 39 million exams performed each year. While breast cancer screening has been one of the most studied medical imaging applications of artificial intelligence, the development and evaluation of the algorithms are hindered due to the lack of well-annotated large-scale publi...

Weakly supervised disease classification of CT imaging suffers from poor localization owing to case-level annotations, where even a positive scan can hold hundreds to thousands of negative slices along multiple planes. Furthermore, although deep learning segmentation and classification models extract distinctly unique combinations of anatomical fea...

Purpose: Digital breast tomosynthesis (DBT) is a limited-angle tomographic breast imaging modality that can be used for breast cancer screening in conjunction with full-field digital mammography (FFDM) or synthetic mammography (SM). Currently, there are five commercial DBT systems that have been approved by the U.S. FDA for breast cancer screening...

Most tissue collections of neoplasms are composed of formalin-fixed and paraffin-embedded (FFPE) excised tumor samples used for routine diagnostics. DNA sequencing is becoming increasingly important in cancer research and clinical management; however, it is difficult to accurately sequence DNA from FFPE samples. We developed and validated a new bio...

Objective: This study aims to develop and validate a novel framework, iPhantom, for automated creation of patient-specific phantoms or digital-twins (DT) using patient medical images. The framework is applied to assess radiation dose to radiosensitive organs in CT imaging of individual patients. Method: From patient CT images, iPhantom segments sel...

Background: The incidence of DCIS has steadily increased with growing concerns regarding overtreatment. Active surveillance is a novel treatment strategy that avoids surgical excision, but identifying patients with occult invasive disease who should be excluded from active surveillance is challenging. Predicting upstaging of DCIS to invasive diseas...

We designed a multi-organ, multi-label disease classification algorithm for computed tomography (CT) scans using case-level labels from radiology text reports. A rule-based algorithm extracted 19,255 disease labels from reports of 13,667 body CT scans from 12,092 subjects. A 3D DenseVNet was trained to segment 3 organ systems: lungs/pleura, liver/g...

The accelerating complexity and variety of medical imaging devices and methods have outpaced the ability to evaluate and optimize their design and clinical use. This is a significant and increasing challenge for both scientific investigations and clinical applications. Evaluations would ideally be done using clinical imaging trials. These experimen...

We investigated PET image quantification when using a uniform attenuation coefficient (μ) for attenuation correction (AC) of anthropomorphic density phantoms derived from high-resolution breast CT scans. A breast PET system was modeled with perfect data corrections except for AC. Using uniform μ for AC resulted in quantitative errors roughly propor...

Weakly supervised disease classification of CT imaging suffers from poor localization owing to case-level annotations, where even a positive scan can hold hundreds to thousands of negative slices along multiple planes. Furthermore, although deep learning segmentation and classification models extract distinctly unique combinations of anatomical fea...

Developing machine learning models for radiology requires large-scale imaging data sets with labels for abnormalities, but the process is challenging due to the size and complexity of the data as well as the cost of labeling. We curated and analyzed a chest computed tomography (CT) data set of 36,316 volumes from 20,201 unique patients. This is the...

Rationale and objectives: The purpose of this study is to quantify breast radiologists' performance at predicting occult invasive disease when ductal carcinoma in situ (DCIS) presents as calcifications on mammography and to identify imaging and histopathological features that are associated with radiologists' performance. Materials and methods:...

Generative Adversarial Networks (GANs) have found applications in natural image synthesis and begin to show promises generating synthetic medical images. In many cases, the ability to perform controlled image synthesis using masked priors such as shape and size of organs is desired. However, mask-guided image synthesis is challenging due to the pix...

Objective: The goal of this study is to use adjunctive classes to improve a predictive model whose performance is limited by the common problems of small numbers of primary cases, high feature dimensionality, and poor class separability. Specifically, our clinical task is to use mammographic features to predict whether ductal carcinoma in situ (DC...

Purpose: our goal is to investigate using only case-level labels extracted automatically from radiology reports to construct a multi-organ, multi-disease classifier for CT scans with deep learning methods. Motivation: worldwide a substantial proportion of people suffers from chest-abdomen-pelvis (CAP). Despite the recent advances, very few studies...

Objective: The purpose of this study was to test the hypothesis whether two-view wide-angle digital breast tomosynthesis (DBT) can replace full-field digital mammography (FFDM) for breast cancer detection. Subjects and methods: In a multireader multicase study, bilateral two-view FFDM and bilateral two-view wide-angle DBT images were independent...

Purpose: The advent of three-dimensional breast imaging systems such as digital breast tomosynthesis (DBT) has great promise for improving the detection and diagnosis of breast cancer. With these new technologies comes an essential need for testing methods to assess the resultant image quality. Although randomized clinical trials are the gold stan...

Background Most ductal carcinoma in situ (DCIS) lesions are first detected on screening mammograms as calcifications. However, false-positive biopsy rates for calcifications range from 30% to 87%. Improved methods to differentiate benign from malignant calcifications are thus needed. Purpose To quantify the growth rates of DCIS and benign breast di...

Anthropomorphic breast phantoms mimic patient anatomy in order to evaluate clinical mammography and digital breast tomosynthesis system performance. Our goal is to create a modular phantom with an anthropomorphic region to allow for improved lesion and calcification detection as well as a uniform region to evaluate standard quality control (QC) met...

This project aims to decrease human burden of labeling CT scans by developing a model that identifies abnormalities within the plain-text-based reports and can be further used as a method to create labels for CT-based models. We propose an iterative method that consists on training a machine learning model to classify CT reports by abnormalities in...

Our goal is to investigate using only case-level labels extracted from radiology reports to construct a multi-disease classifier for CT scans. We chose three lung diseases: atelectasis, edema, nodule and pneumonia. From a dataset of approximately 5,000 chest CT cases from our institution, we used a rule-based model to analyze those radiologist repo...

Our goal is to develop a 2.5D CNN model to detect multiple diseases in multiple organs in CT scans. To establish feasibility, in this study we investigated detection of atelectasis in the lungs. Our hypothesis is that by using information from all of the three views (coronal, sagittal and axial), we may achieve a better classification result, becau...

When conducting machine learning algorithms on classification and detection of abnormalities for medical imaging, many researchers are faced with the problem that it is hard to get enough labeled data. To solve this problem, we plan to use machine learning algorithms to identify abnormalities within existing radiologist reports, thus creating case-...

Background: Deep learning, especially deep convolutional neural network (CNN), has emerged as a promising approach for many image recognition or classification tasks, demonstrating human or even superhuman performance. Used as feature extractor, some pre-trained CNN models can match or surpass the performance of domain-specific, “handcrafted” featu...

Purpose: The aim of this study was to determine whether deep features extracted from digital mammograms using a pretrained deep convolutional neural network are prognostic of occult invasive disease for patients with ductal carcinoma in situ (DCIS) on core needle biopsy. Methods: In this retrospective study, digital mammographic magnification vi...

Digital breast tomosynthesis (DBT) acquires a series of projection images from different angles as an x-ray source rotates around the breast. Such imaging geometry lends DBT naturally to stereoscopic viewing as two projection images with a reasonable separation angle can easily form a stereo pair. This simulation study assessed the efficacy of ster...

Purpose: The limited number of 3D patient-based breast phantoms available could be augmented by synthetic breast phantoms in order to facilitate virtual clinical trials using model observers for breast imaging optimization and evaluation. Methods: These synthetic breast phantoms were developed using Principal Component Analysis (PCA) to reduce t...

Rationale and objectives: This study aimed to determine whether mammographic features assessed by radiologists and using computer algorithms are prognostic of occult invasive disease for patients showing ductal carcinoma in situ (DCIS) only in core biopsy. Materials and methods: In this retrospective study, we analyzed data from 99 subjects with...

Purpose: Physical phantoms are central to the evaluation of 2D and 3D breast imaging systems. Currently available physical phantoms have limitations including unrealistic uniform background structure, large expense, or excessive fabrication time. The purpose of this work is to outline a method for rapidly creating realistic, inexpensive physical a...

Purpose:The purpose of this study is to quantify the differences in detectability between full field digital mammography (FFDM), digital breast tomosynthesis (DBT), and synthetic mammography (SM) for challenging, low contrast signals, in the context of both a uniform and an anthropomorphic, textured phantom. Methods:Images of the phantoms were acqu...

This study aims to characterize the effect of background tissue density and heterogeneity on the detection of irregular masses in breast tomosynthesis, while demonstrating the capability of the sophisticated tools that can be used in the design, implementation, and performance analysis of virtual clinical trials (VCTs). Twenty breast phantoms from...

Purpose: The authors are developing a series of computational breast phantoms based on breast CT data for imaging research. In this work, the authors develop a program that will allow a user to alter the phantoms to simulate the effect of gravity and compression of the breast (craniocaudal or mediolateral oblique) making the phantoms applicable to...

Purpose: In medical imaging systems, proper rendition of anatomy is essential in discerning normal tissue from disease. Currently, digital breast tomosynthesis (DBT) systems are evaluated using subjective evaluation of lesion visibility in uniform phantoms. This study involved the development of a new methodology to objectively measure the renditi...

Although digital mammography has reduced breast cancer mortality by approximately 30%, sensitivity and specificity are still far from perfect. In particular, the performance of mammography is especially limited for women with dense breast tissue. Two out of every three biopsies performed in the U.S. are unnecessary, thereby resulting in increased p...

Digital breast tomosynthesis (DBT) can improve lesion visibility by eliminating the issue of overlapping breast tissue present in mammography. However, this new modality likely requires new approaches to training. The issue of training in DBT is not well explored. We propose a computer-aided educational approach for DBT training. Our hypothesis is...

To facilitate rigorous virtual clinical trials using model observers for breast imaging optimization and evaluation, we demonstrated a method of defining statistical models, based on 177 sets of breast CT patient data, in order to generate tens of thousands of unique digital breast phantoms. In order to separate anatomical texture from variation in...