Stefan Jaeger

Efficient feature extraction and likelihood fusion for vehicle tracking in low frame rate airborne video

Very large format video or wide-area motion imagery (WAMI) acquired by an airborne camera sensor array is characterized by persistent observation over a large field-of-view with high spatial resolution but low frame rates (i.e. one to ten frames per second). Current WAMI sensors have sufficient coverage and resolution to track vehicles for many hours using just a single airborne platform. We have developed an interactive low frame rate tracking system based on a derived rich set of features for vehicle detection using appearance modeling combined with saliency estimation and motion prediction. Instead of applying subspace methods to very high-dimensional feature vectors, we tested the performance of feature fusion to locate the target of interest within the prediction window. Preliminary results show that fusing the feature likelihood maps improves detection but fusing feature maps combined with saliency information actually degrades performance.

Lung Segmentation in Chest Radiographs Using Anatomical Atlases With Nonrigid Registration

The National Library of Medicine (NLM) is developing a digital chest X-ray (CXR) screening system for deployment in resource constrained communities and developing countries worldwide with a focus on early detection of tuberculosis. A critical component in the computer-aided diagnosis of digital CXRs is the automatic detection of the lung regions. In this paper, we present a nonrigid registration-driven robust lung segmentation method using image retrieval-based patient specific adaptive lung models that detects lung boundaries, surpassing state-of-the-art performance. The method consists of three main stages: 1) a content-based image retrieval approach for identifying training images (with masks) most similar to the patient CXR using a partial Radon transform and Bhattacharyya shape similarity measure, 2) creating the initial patient-specific anatomical model of lung shape using SIFT-flow for deformable registration of training masks to the patient CXR, and 3) extracting refined lung boundaries using a graph cuts optimization approach with a customized energy function. Our average accuracy of 95.4% on the public JSRT database is the highest among published results. A similar degree of accuracy of 94.1% and 91.7% on two new CXR datasets from Montgomery County, MD, USA, and India, respectively, demonstrates the robustness of our lung segmentation approach.

Automatic Tuberculosis Screening Using Chest Radiographs

Tuberculosis is a major health threat in many regions of the world. Opportunistic infections in immunocompromised HIV/AIDS patients and multi-drug-resistant bacterial strains have exacerbated the problem, while diagnosing tuberculosis still remains a challenge. When left undiagnosed and thus untreated, mortality rates of patients with tuberculosis are high. Standard diagnostics still rely on methods developed in the last century. They are slow and often unreliable. In an effort to reduce the burden of the disease, this paper presents our automated approach for detecting tuberculosis in conventional posteroanterior chest radiographs. We first extract the lung region using a graph cut segmentation method. For this lung region, we compute a set of texture and shape features, which enable the X-rays to be classified as normal or abnormal using a binary classifier. We measure the performance of our system on two datasets: a set collected by the tuberculosis control program of our local countys health department in the United States, and a set collected by Shenzhen Hospital, China. The proposed computer-aided diagnostic system for TB screening, which is ready for field deployment, achieves a performance that approaches the performance of human experts. We achieve an area under the ROC curve (AUC) of 87% (78.3% accuracy) for the first set, and an AUC of 90% (84% accuracy) for the second set. For the first set, we compare our system performance with the performance of radiologists. When trying not to miss any positive cases, radiologists achieve an accuracy of about 82% on this set, and their false positive rate is about half of our systems rate.

Automatic screening for tuberculosis in chest radiographs: a survey

Tuberculosis (TB) is a major global health threat. An estimated one-third of the worlds population has a history of TB infection, and millions of new infections are occurring every year. The advent of new powerful hardware and software techniques has triggered attempts to develop computer-aided diagnostic systems for TB detection in support of inexpensive mass screening in developing countries. In this paper, we describe the medical background of TB detection in chest X-rays and present a survey of the recent approaches using computer-aided detection. After a thorough research of the computer science literature for such systems or related methods, we were able to identify 16 papers, including our own, written between 1996 and early 2013. These papers show that TB screening is a challenging task and an open research problem. We report on the progress to date and describe experimental screening systems that have been developed.

Detecting tuberculosis in radiographs using combined lung masks

Tuberculosis (TB) is a major health threat in many regions of the world, while diagnosing tuberculosis still remains a challenge. Mortality rates of patients with undiagnosed TB are high. Modern diagnostic techniques are often too slow or too expensive for highly-populated developing countries that bear the brunt of the disease. In an effort to reduce the burden of the disease, this paper presents an automated approach for detecting TB on conventional posteroanterior chest radiographs. The idea is to provide developing countries, which have limited access to radiological services and radiological expertise, with an inexpensive detection system that allows screening of large parts of the population in rural areas. In this paper, we present results produced by our TB screening system. We combine a lung shape model, a segmentation mask, and a simple intensity model to achieve a better segmentation mask for the lung. With the improved masks, we achieve an area under the ROC curve of more than 83%, measured on data compiled within a tuberculosis control program.

Two public chest X-ray datasets for computer-aided screening of pulmonary diseases

The U.S. National Library of Medicine has made two datasets of postero-anterior (PA) chest radiographs available to foster research in computer-aided diagnosis of pulmonary diseases with a special focus on pulmonary tuberculosis (TB). The radiographs were acquired from the Department of Health and Human Services, Montgomery County, Maryland, USA and Shenzhen No. 3 Peoples Hospital in China. Both datasets contain normal and abnormal chest X-rays with manifestations of TB and include associated radiologist readings.

CNN-based image analysis for malaria diagnosis

Malaria is a major global health threat. The standard way of diagnosing malaria is by visually examining blood smears for parasite-infected red blood cells under the microscope by qualified technicians. This method is inefficient and the diagnosis depends on the experience and the knowledge of the person doing the examination. Automatic image recognition technologies based on machine learning have been applied to malaria blood smears for diagnosis before. However, the practical performance has not been sufficient so far. This study proposes a new and robust machine learning model based on a convolutional neural network (CNN) to automatically classify single cells in thin blood smears on standard microscope slides as either infected or uninfected. In a ten-fold cross-validation based on 27,578 single cell images, the average accuracy of our new 16-layer CNN model is 97.37%. A transfer learning model only achieves 91.99% on the same images. The CNN model shows superiority over the transfer learning model in all performance indicators such as sensitivity (96.99% vs 89.00%), specificity (97.75% vs 94.98%), precision (97.73% vs 95.12%), F1 score (97.36% vs 90.24%), and Matthews correlation coefficient (94.75% vs 85.25%).

Classification of cell cycle phases in 3D confocal microscopy using PCNA and chromocenter features

Cell cycle progression studies using subcellular markers offer important insight into cellular mechanisms of disease and therapeutic drug development. Due to the large volumes of microscopy data involved in such studies, a manual approach to extracting quantitative information is not only prohibitive but error prone. We present an automatic cell cycle phase identification algorithm applied to 3D spinning disk confocal microscopy imagery of mouse embryonic fibroblast cells. In our training dataset, each 3D image stack depicts a single cell in a manually identified cell phase, and is recorded via two channels showing the fluorescently marked protein PCNA and the chromocenters, respectively. We use a 3D k-means approach to segment each volume and extract a set of shape and curvature features to characterize the subcellular foci patterns associated with cell cycle phases for each channel. Radial features are used to describe the spatial distribution of PCNA over the course of the cell cycle. A support vector machine (SVM) classifier using 234 features was trained and achieved a recognition rate of 83% for the chromocenter and 86% for the PCNA channels separately on the testing data. A combined SVM classifier using both channels and 468 features further improved the accuracy to nearly 92% for five phases (G1, SE, SM, SL, G2) and shows promising scalability.

Tuberculosis screening of chest radiographs

Tuberculosis (TB) is one of the most common causes of death by an infectious agent,1 with an estimated nine million new cases appearing every year. About one-third of the world’s population is infected with Mycobacterium tuberculosis, the bacterial strain that causes the majority of cases. TB is most prevalent in sub-Saharan Africa and Southeast Asia, where widespread poverty and malnutrition reduce resistance to the disease. Despite progress made in prevention, diagnosis, and treatment, the emergence of multi-drug-resistant bacterial strains and opportunistic infections in immunocompromised patients, for example, those with HIV (human immunodeficiency virus), has exacerbated the problem. However, the likelihood of curing TB is improved when it is diagnosed at an early stage. Computer-aided screening and diagnosis have received increasing attention with the advent of digital chest x-rays (CXRs), which allow image processing that traditional film x-rays do not. Here, we describe our progress—in collaboration with the Academic Model Providing Access to Healthcare (AMPATH)—toward improving TB diagnosis with intelligent software designed for portable scanners that can easily be used in remote locations. Several skin tests are available—based on immune response— for determining whether an individual has been exposed to TB. However, it is often not possible to make a final diagnosis based on these tests alone. Thus, physicians typically use additional tests—including posteroanterior x-rays of the patient’s chest—to confirm or rule out infection. The abnormalities displayed in radiographs in TB are generally diffuse, and the distinction between normal anatomical structures and abnormal patterns is difficult to determine. Furthermore, low-contrast images complicate identification of the subtle radiographic manifestations of TB. The lack of adequate radiological services in the worst-affected areas necessitates an automated, Figure 1. Contrast-enhanced chest x-ray, normalized with a histogramequalization technique, showing a nodule in the right lower lobe.2

Peroxiredoxin 6 (Prdx6) supports NADPH oxidase1 (Nox1)-based superoxide generation and cell migration.

Nox1 is an abundant source of reactive oxygen species (ROS) in colon epithelium recently shown to function in wound healing and epithelial homeostasis. We identified Peroxiredoxin 6 (Prdx6) as a novel binding partner of Nox activator 1 (Noxa1) in yeast two-hybrid screening experiments using the Noxa1 SH3 domain as bait. Prdx6 is a unique member of the Prdx antioxidant enzyme family exhibiting both glutathione peroxidase and phospholipase A2 activities. We confirmed this interaction in cells overexpressing both proteins, showing Prdx6 binds to and stabilizes wild type Noxa1, but not the SH3 domain mutant form, Noxa1 W436R. We demonstrated in several cell models that Prdx6 knockdown suppresses Nox1 activity, whereas enhanced Prdx6 expression supports higher Nox1-derived superoxide production. Both peroxidase- and lipase-deficient mutant forms of Prdx6 (Prdx6 C47S and S32A, respectively) failed to bind to or stabilize Nox1 components or support Nox1-mediated superoxide generation. Furthermore, the transition-state substrate analogue inhibitor of Prdx6 phospholipase A2 activity (MJ-33) was shown to suppress Nox1 activity, suggesting Nox1 activity is regulated by the phospholipase activity of Prdx6. Finally, wild type Prdx6, but not lipase or peroxidase mutant forms, supports Nox1-mediated cell migration in the HCT-116 colon epithelial cell model of wound closure. These findings highlight a novel pathway in which this antioxidant enzyme positively regulates an oxidant-generating system to support cell migration and wound healing.