Removing batch effects from histopathological images for enhanced cancer diagnosis.

Researchers have developed computer-aided decision support systems for translational medicine that aim to objectively and efficiently diagnose cancer using histopathological images. However, the performance of such systems is confounded by nonbiological experimental variations or "batch effects" that can commonly occur in histopathological data, especially when images are acquired using different imaging devices and patient samples. This is even more problematic in large-scale studies in which cross-laboratory sharing of large volumes of data is necessary. Batch effects can change quantitative morphological image features and decrease the prediction performance. Using four batches of renal tumor images, we compare one image-level and five feature-level batch effect removal methods. Principal component variation analysis shows that batch is a large source of variance in image features. Results show that feature-level normalization methods reduce batch-contributed variance to almost zero. Moreover, feature-level normalization, especially ComBatN, improves cross-batch and combined-batch prediction performance. Compared to no normalization, ComBatN improves performance in 83% and 90% of cross-batch and combined-batch prediction models, respectively.

Kinect-based rehabilitation system for patients with traumatic brain injury.

Traumatic brain injury (TBI) is one of the leading causes of death and disability in the age group of 0 - 44 years. Though physical exercises have proven benefits in the rehabilitation process, the compliance rate of patients in the home environment is poor. In this paper we propose a system, MotionTalk, which captures and analyses motion data acquired using Microsoft Kinect. It is designed to give a real time quantitative assessment of the exercises performed by TBI patients at home with respect to the same exercises performed in the clinic by utilizing relatively inexpensive contactless sensing and dynamic programming techniques. In comparison to previous reminder systems, wearable systems, and motion capture systems, MotionTalk is less intrusive in nature and inexpensive to deploy at home because it is based on readily available hardware. This was developed and tested on able bodied volunteers and in the future we hope to test it on patients with TBI, after IRB approval.

iACT--an interactive mHealth monitoring system to enhance psychotherapy for adolescents with sickle cell disease.

Sickle cell disease (SCD) is the most common inherited disease, and SCD symptoms impact functioning and well-being. For example, adolescents with SCD have a higher tendency of psychological problems than the general population. Acceptance and Commitment Therapy (ACT), a cognitive-behavioral therapy, is an effective intervention to promote quality of life and functioning in adolescents with chronic illness. However, traditional visit-based therapy sessions are restrained by challenges, such as limited follow-up, insufficient data collection, low treatment adherence, and delayed intervention. In this paper, we present Instant Acceptance and Commitment Therapy (iACT), a system designed to enhance the quality of pediatric ACT. iACT utilizes text messaging technology, which is the most popular cell phone activity among adolescents, to conduct real-time psychotherapy interventions. The system is built on cloud computing technologies, which provides a convenient and cost-effective monitoring environment. To evaluate iACT, a trial with 60 adolescents with SCD is being conducted in conjunction with the Georgia Institute of Technology, Children's Healthcare of Atlanta, and Georgia State University.

A pilot biomedical engineering course in rapid prototyping for mobile health.

Rapid prototyping of medically assistive mobile devices promises to fuel innovation and provides opportunity for hands-on engineering training in biomedical engineering curricula. This paper presents the design and outcomes of a course offered during a 16-week semester in Fall 2011 with 11 students enrolled. The syllabus covered a mobile health design process from end-to-end, including storyboarding, non-functional prototypes, integrated circuit programming, 3D modeling, 3D printing, cloud computing database programming, and developing patient engagement through animated videos describing the benefits of a new device. Most technologies presented in this class are open source and thus provide unlimited "hackability". They are also cost-effective and easily transferrable to other departments.

Unmixing of spectrally similar quantum dots using filter selection.

This paper explores the possibilities for quantitative analysis of multiplexed Quantum Dot Immunohistochemical (QDIHC) staining using a 10-slot fluorescence microscope filter wheel. QDs are an ideal fluorophore for staining biomarkers due to their unique properties, including greater photostability and relatively narrower emission bandwidths compared to organic dyes. We imaged a slide containing 5 pure QD spots and 6 QD mixtures with a customized scanning fluorescence microscope. The QD mixtures contained either two or three QDs in equal amounts. Ten filter cubes were used to gather emission signal and then fast non-negative least squares regression (FNNLS) performed the unmixing process by assigning components of the 10-channel raw data to one of the five QDs used. the average error in the unmixing process was measured to be 7.60% when all filters were used and 7.80% when only 6 filters were used.

Pathology imaging informatics for quantitative analysis of whole-slide images.

OBJECTIVES: With the objective of bringing clinical decision support systems to reality, this article reviews histopathological whole-slide imaging informatics methods, associated challenges, and future research opportunities. TARGET AUDIENCE: This review targets pathologists and informaticians who have a limited understanding of the key aspects of whole-slide image (WSI) analysis and/or a limited knowledge of state-of-the-art technologies and analysis methods. SCOPE: First, we discuss the importance of imaging informatics in pathology and highlight the challenges posed by histopathological WSI. Next, we provide a thorough review of current methods for: quality control of histopathological images; feature extraction that captures image properties at the pixel, object, and semantic levels; predictive modeling that utilizes image features for diagnostic or prognostic applications; and data and information visualization that explores WSI for de novo discovery. In addition, we highlight future research directions and discuss the impact of large public repositories of histopathological data, such as the Cancer Genome Atlas, on the field of pathology informatics. Following the review, we present a case study to illustrate a clinical decision support system that begins with quality control and ends with predictive modeling for several cancer endpoints. Currently, state-of-the-art software tools only provide limited image processing capabilities instead of complete data analysis for clinical decision-making. We aim to inspire researchers to conduct more research in pathology imaging informatics so that clinical decision support can become a reality.

Semiconductor quantum dots for bioimaging and biodiagnostic applications.

Semiconductor quantum dots (QDs) are light-emitting particles on the nanometer scale that have emerged as a new class of fluorescent labels for chemical analysis, molecular imaging, and biomedical diagnostics. Compared with traditional fluorescent probes, QDs have unique optical and electronic properties such as size-tunable light emission, narrow and symmetric emission spectra, and broad absorption spectra that enable the simultaneous excitation of multiple fluorescence colors. QDs are also considerably brighter and more resistant to photobleaching than are organic dyes and fluorescent proteins. These properties are well suited for dynamic imaging at the single-molecule level and for multiplexed biomedical diagnostics at ultrahigh sensitivity. Here, we discuss the fundamental properties of QDs; the development of next-generation QDs; and their applications in bioanalytical chemistry, dynamic cellular imaging, and medical diagnostics. For in vivo and clinical imaging, the potential toxicity of QDs remains a major concern. However, the toxic nature of cadmium-containing QDs is no longer a factor for in vitro diagnostics, so the use of multicolor QDs for molecular diagnostics and pathology is probably the most important and clinically relevant application for semiconductor QDs in the immediate future.

Life sciences domain analysis model.

OBJECTIVE: Meaningful exchange of information is a fundamental challenge in collaborative biomedical research. To help address this, the authors developed the Life Sciences Domain Analysis Model (LS DAM), an information model that provides a framework for communication among domain experts and technical teams developing information systems to support biomedical research. The LS DAM is harmonized with the Biomedical Research Integrated Domain Group (BRIDG) model of protocol-driven clinical research. Together, these models can facilitate data exchange for translational research. MATERIALS AND METHODS: The content of the LS DAM was driven by analysis of life sciences and translational research scenarios and the concepts in the model are derived from existing information models, reference models and data exchange formats. The model is represented in the Unified Modeling Language and uses ISO 21090 data types. RESULTS: The LS DAM v2.2.1 is comprised of 130 classes and covers several core areas including Experiment, Molecular Biology, Molecular Databases and Specimen. Nearly half of these classes originate from the BRIDG model, emphasizing the semantic harmonization between these models. Validation of the LS DAM against independently derived information models, research scenarios and reference databases supports its general applicability to represent life sciences research. DISCUSSION: The LS DAM provides unambiguous definitions for concepts required to describe life sciences research. The processes established to achieve consensus among domain experts will be applied in future iterations and may be broadly applicable to other standardization efforts. CONCLUSIONS: The LS DAM provides common semantics for life sciences research. Through harmonization with BRIDG, it promotes interoperability in translational science.

Activity and school attendance monitoring system for adolescents with sickle cell disease.

Sickle cell disease, the most common hemoglobin disorder, affects major organ systems with symptoms of pain, anemia and a multitude of chronic conditions. For adolescents, the disease adversely affects school attendance, academic progress and social activity. To effectively study the relationship among school attendance and other factors like demographics and academic performance, studies have relied on self-reporting and school records, all of which have some bias. In this study we design and prototype a system, called SickleSAM (Sickle cell School attendance and Activity Monitoring system), for automatically monitoring school attendance and daily activity of adolescents with sickle cell disease. SickleSAM intends to remove human bias and inaccuracies. The system uses built-in GPS to collect data which will be recorded into a cloud database using Short Messaging Service technology. SickleSAM is developed by Georgia Institute of Technology in conjunction with Children's Healthcare of Atlanta (CHOA). System effectiveness is being evaluated using a trial of 10 adolescents with the disease.

SickleREMOTE: A Two-Way Text Messaging System for Pediatric Sickle Cell Disease Patients.

Sickle cell disease, the most common hemoglobinopathy in the world, affects patient lives from early childhood. Effective care of sickle cell disease requires frequent medical monitoring, such as tracking the frequency, severity, and duration of painful events. Conventional monitoring includes paper- or web-based reporting diaries. These systems require that patients carry forms, which are easily lost, or laptop computers, which are impractical to scale to large populations. Both are prone to sporadic use by older adolescents due to lack of reminders. In this paper, we design and prototype a Sickle cell disease REporting and MOnitoring TElemedicine system (SickleREMOTE), aiming to resolve limitations of conventional monitoring diaries. This monitoring system is configured as automated short message service text (SMS-text) messages that arrive at a mobile phone anywhere on a cellular network. The messages may be reminders to encourage treatment adherence or questionnaires to collect self-assessed clinical data relating to treatment adjustments. Patients respond to the messages using pre-determined templates and a cloud database parses and stores messages automatically. Providers use a web-based interface to view, analyze, and download collected data. SickleREMOTE is developed by Georgia Institute of Technology in conjunction with Children's Healthcare of Atlanta (CHOA). System effectiveness will be evaluated using a trial of 30 adolescents with sickle cell disease and measured by response rate, time to response, error rate, and correspondence with data collected by telephone calls.

Feasibility of multiplex quantum dot stain using primary antibodies from four distinct host animals.

We discuss the feasibility of multiplex QD stain for four biomarkers and our progress in finding four suitable biomarkers from four different hosts. There is a demand for using more than three fluorescent probes on a single tissue sample for disease detection to offer a more reliable prediction of disease progression. We developed a protocol for targeting four biomarkers using four primary antibodies from four different animal hosts. We performed primary-secondary antibody binding assays on nitrocellulose paper and stained breast cancer microarray slides with known expression of ER, PR, and HER2. We identified the lack of a standard protocol and the limited supply of primary antibodies from hosts other than rabbit and mouse in the market as key challenges. The results show variable success in both assays, but indicate future potential for this protocol with more development.

caCORRECT2: Improving the accuracy and reliability of microarray data in the presence of artifacts.

BACKGROUND: In previous work, we reported the development of caCORRECT, a novel microarray quality control system built to identify and correct spatial artifacts commonly found on Affymetrix arrays. We have made recent improvements to caCORRECT, including the development of a model-based data-replacement strategy and integration with typical microarray workflows via caCORRECT's web portal and caBIG grid services. In this report, we demonstrate that caCORRECT improves the reproducibility and reliability of experimental results across several common Affymetrix microarray platforms. caCORRECT represents an advance over state-of-art quality control methods such as Harshlighting, and acts to improve gene expression calculation techniques such as PLIER, RMA and MAS5.0, because it incorporates spatial information into outlier detection as well as outlier information into probe normalization. The ability of caCORRECT to recover accurate gene expressions from low quality probe intensity data is assessed using a combination of real and synthetic artifacts with PCR follow-up confirmation and the affycomp spike in data. The caCORRECT tool can be accessed at the website: http://cacorrect.bme.gatech.edu. RESULTS: We demonstrate that (1) caCORRECT's artifact-aware normalization avoids the undesirable global data warping that happens when any damaged chips are processed without caCORRECT; (2) When used upstream of RMA, PLIER, or MAS5.0, the data imputation of caCORRECT generally improves the accuracy of microarray gene expression in the presence of artifacts more than using Harshlighting or not using any quality control; (3) Biomarkers selected from artifactual microarray data which have undergone the quality control procedures of caCORRECT are more likely to be reliable, as shown by both spike in and PCR validation experiments. Finally, we present a case study of the use of caCORRECT to reliably identify biomarkers for renal cell carcinoma, yielding two diagnostic biomarkers with potential clinical utility, PRKAB1 and NNMT. CONCLUSIONS: caCORRECT is shown to improve the accuracy of gene expression, and the reproducibility of experimental results in clinical application. This study suggests that caCORRECT will be useful to clean up possible artifacts in new as well as archived microarray data.

Informatics and standards for nanomedicine technology.

There are several issues to be addressed concerning the management and effective use of information (or data), generated from nanotechnology studies in biomedical research and medicine. These data are large in volume, diverse in content, and are beset with gaps and ambiguities in the description and characterization of nanomaterials. In this work, we have reviewed three areas of nanomedicine informatics: information resources; taxonomies, controlled vocabularies, and ontologies; and information standards. Informatics methods and standards in each of these areas are critical for enabling collaboration; data sharing; unambiguous representation and interpretation of data; semantic (meaningful) search and integration of data; and for ensuring data quality, reliability, and reproducibility. In particular, we have considered four types of information standards in this article, which are standard characterization protocols, common terminology standards, minimum information standards, and standard data communication (exchange) formats. Currently, because of gaps and ambiguities in the data, it is also difficult to apply computational methods and machine learning techniques to analyze, interpret, and recognize patterns in data that are high dimensional in nature, and also to relate variations in nanomaterial properties to variations in their chemical composition, synthesis, characterization protocols, and so on. Progress toward resolving the issues of information management in nanomedicine using informatics methods and standards discussed in this article will be essential to the rapidly growing field of nanomedicine informatics.

Feasibility analysis of high resolution tissue image registration using 3-D synthetic data.

BACKGROUND: Registration of high-resolution tissue images is a critical step in the 3D analysis of protein expression. Because the distance between images (~4-5µm thickness of a tissue section) is nearly the size of the objects of interest (~10-20µm cancer cell nucleus), a given object is often not present in both of two adjacent images. Without consistent correspondence of objects between images, registration becomes a difficult task. This work assesses the feasibility of current registration techniques for such images. METHODS: We generated high resolution synthetic 3-D image data sets emulating the constraints in real data. We applied multiple registration methods to the synthetic image data sets and assessed the registration performance of three techniques (i.e., mutual information (MI), kernel density estimate (KDE) method [1], and principal component analysis (PCA)) at various slice thicknesses (with increments of 1µm) in order to quantify the limitations of each method. RESULTS: Our analysis shows that PCA, when combined with the KDE method based on nuclei centers, aligns images corresponding to 5µm thick sections with acceptable accuracy. We also note that registration error increases rapidly with increasing distance between images, and that the choice of feature points which are conserved between slices improves performance. CONCLUSIONS: We used simulation to help select appropriate features and methods for image registration by estimating best-case-scenario errors for given data constraints in histological images. The results of this study suggest that much of the difficulty of stained tissue registration can be reduced to the problem of accurately identifying feature points, such as the center of nuclei.

caREMOTE: the design of a cancer reporting and monitoring telemedicine system for domestic care.

After receiving cancer treatment, patients often experience a decline of HRQoL (health-related quality of life). Physicians typically evaluate HRQoL during periodic clinical visits. However, out-patient reporting of vital signals between two visits could be used to interpret the decline of HRQoL. Considering that the vast majority of patients recovering from cancer are not in hospitals, it is often impractical for the care providers to collect these data. In this paper, we design and prototype caREMOTE, a cancer reporting and monitoring telemedicine system, which can be used in domestic cancer care. By extending a standard clinical trial informatics model, we build a prototype on cloud computing services that can be accessed by a mobile application. We aim to maximize the potential of caREMOTE to help medical practitioners efficiently monitor discharged patients' HRQoL and vital signals, and facilitate data reusability and system interoperability in future collaborative cancer research.

TissueWikiMobile: an Integrative Protein Expression Image Browser for Pathological Knowledge Sharing and Annotation on a Mobile Device.

Doctors need fast and convenient access to medical data. This motivates the use of mobile devices for knowledge retrieval and sharing. We have developed TissueWikiMobile on the Apple iPhone and iPad to seamlessly access TissueWiki, an enormous repository of medical histology images. TissueWiki is a three terabyte database of antibody information and histology images from the Human Protein Atlas (HPA). Using TissueWikiMobile, users are capable of extracting knowledge from protein expression, adding annotations to highlight regions of interest on images, and sharing their professional insight. By providing an intuitive human computer interface, users can efficiently operate TissueWikiMobile to access important biomedical data without losing mobility. TissueWikiMobile furnishes the health community a ubiquitous way to collaborate and share their expert opinions not only on the performance of various antibodies stains but also on histology image annotation.

SimplevisGrid: grid services for visualization of diverse biomedical knowledge and molecular systems data.

Biomedical data visualization is a great challenge due to the scale, complexity, and diversity of systems, system component interactions and experimental data. Standards for interoperable data are a good start to addressing these problems, but standardization of visualization technologies is an emerging topic. SimpleVisGrid builds on Cancer Biomedical Informatics Grid (caBIG) common infrastructure for cancer research, and clearly specifies and extends three standard data formats for inputs and outputs to grid services: comma-separated values (CSV), Portable Network Graphics (PNG), and Scalable Vector Graphics (SVG). Four prototype visualizations are available: 2D array data quality visualization, correlation heatmaps between high-dimensional data and associated meta-data, feature landscapes, and biochemical or semantic network graphs. The services and data model are prepared for submission for caBIG Silver-level compatibility review and for integration into automated research workflows. Making these tools available to caBIG developers and ultimately to biomedical researchers can (1) help with biomedical communication, discovery, and decision-making, (2) encourage more research on standardization of visualization formats, and (3) improve the efficiency of large data transfers across the grid.

Development of an automatic quantification method for cancer tissue microarray study.

Clinical Histopathology is based on the analysis of immunohistochemistry (IHC) stained tissue images. Selection of antibodies for detecting the presence, type, and grade of cancerous tissue has a great influence on the diagnostic potential of IHC tests. Automated evaluation methods for tissue microarrays applied to many combinations of antibody and tissue type can speed development of new clinical assays. We present an automatic method that successfully quantifies stain intensity, fraction of cells stained and sub-cellular location of staining in tissue microarray images. The method combines an opponent color preprocessor and a novel statistical approach for identifying brown and blue staining, followed by multilevel morphological processing. We verify the capability of our method by comparing the results to manually annotated image databases. We also demonstrate cross-tissue robustness using two clinical case study data.

Diagnostic biomarkers for renal cell carcinoma: selection using novel bioinformatics systems for microarray data analysis.

The differential diagnosis of clear cell, papillary, and chromophobe renal cell carcinoma is clinically important, because these tumor subtypes are associated with different pathobiology and clinical behavior. For cases in which histopathology is equivocal, immunohistochemistry and quantitative reverse transcriptase-polymerase chain reaction can assist in the differential diagnosis by measuring expression of subtype-specific biomarkers. Several renal tumor biomarkers have been discovered in expression microarray studies. However, due to heterogeneity of gene and protein expression, additional biomarkers are needed for reliable diagnostic classification. We developed novel bioinformatics systems to identify candidate renal tumor biomarkers from the microarray profiles of 45 clear cell, 16 papillary, and 10 chromophobe renal cell carcinomas; the microarray data was derived from 2 independent published studies. The ArrayWiki biocomputing system merged the microarray data sets into a single file, so gene expression could be analyzed from a larger number of tumors. The caCORRECT system removed non-random sources of error from the microarray data, and the omniBioMarker system analyzed data with several gene-ranking algorithms to identify algorithms effective at recognizing previously described renal tumor biomarkers. We predicted these algorithms would also be effective at identifying unknown biomarkers that could be verified by independent methods. We selected 6 novel candidate biomarkers from the omniBioMarker analysis and verified their differential expression in formalin-fixed paraffin-embedded tissues by quantitative reverse transcriptase-polymerase chain reaction and immunohistochemistry. The candidate biomarkers were carbonic anhydrase IX, ceruloplasmin, schwannomin-interacting protein 1, E74-like factor 3, cytochrome c oxidase subunit 5a, and acetyl-CoA acetyltransferase 1. Quantitative reverse transcriptase-polymerase chain reaction was performed on 17 clear cell, 13 papillary and 7 chromophobe renal cell carcinoma. Carbonic anhydrase IX and ceruloplasmin were overexpressed in clear cell renal cell carcinoma; schwannomin-interacting protein 1 and E74-like factor 3 were overexpressed in papillary renal cell carcinoma; and cytochrome c oxidase subunit 5a and acetyl-CoA acetyltransferase 1 were overexpressed in chromophobe renal cell carcinoma. Immunohistochemistry was performed on tissue microarrays containing 66 clear cell, 16 papillary, and 12 chromophobe renal cell carcinomas. Cytoplasmic carbonic anhydrase IX staining was significantly associated with clear cell renal cell carcinoma. Strong cytoplasmic schwannomin-interacting protein 1 and cytochrome c oxidase subunit 5a staining were significantly more frequent in papillary and chromophobe renal cell carcinoma, respectively. In summary, we developed a novel process for identifying candidate renal tumor biomarkers from microarray data, and verifying differential expression in independent assays. The tumor biomarkers have potential utility as a multiplex expression panel for classifying renal cell carcinoma with equivocal histology. Biomarker expression assays are increasingly important for renal cell carcinoma diagnosis, as needle core biopsies become more common and different therapies for tumor subtypes continue to be developed.