This paper presents findings from a two-month comparative study involving a total of 36 participants using a social fitness application called HealthyTogether. Our aim was to understand whether and how patients with chronic diseases and healthy adults respond differently to social incentives, such as competition, cooperation, and accountability, and how these incentives could relate to their engagement in physical activities. We found that community leaderboard served different goals: healthy adults mainly used it to compete with others, and patients used it to validate their normalcy. For the patients, pairing up with strong ties fostered fulfilling fitness goals, while exercising with strangers diminished them. This study shows that social fitness application design for patients should take into account their need for support from communities and close relationships. Furthermore, our findings point towards opportunities for leveraging and modifying existing social fitness applications for patients' health management.

We propose near-affine-invariant texture descriptors derived from isotropic wavelet frames for the characterization of lung tissue patterns in high-resolution computed tomography (HRCT) imaging. Affine invariance is desirable to enable learning of nondeterministic textures without a priori localizations, orientations, or sizes. When combined with complementary gray-level histograms, the proposed method allows a global classification accuracy of 76.9% with balanced precision among five classes of lung tissue using a leave-one-patient-out cross validation, in accordance with clinical practice.

This paper concerns lung tissue classification using asymmetric-margin support vector machine (ASVM) to handle the imbalance of the positive and negative classes in a one-against-all multiclass classification problem. The hyperparameters of the algorithm are obtained using an optimization of the upper bound of the leave-one-out error of the ASVM. The ASVM is applied on the dataset with its original distribution and oversampled so that the ratio of the examples is equal to the prevalence of patients having the tissue in the database. The two versions of the ASVM models were compared with a model build with a conventional SVM. The ASVM improved the results obtained with a conventional SVM. The incorporation of prior knowledge concerning the prevalence of the patients improved the results obtained with ASVM.

Today's medical institutions produce enormous amounts of data on patients, including multimedia data, which is increasingly produced in digital form. These data in their clinical context contain much information and experience that is currently not being used up to its full potential. Through the digital form the data has become accessible for automatic analysis and treatment for a variety of applications. At the same time, the variety of images produced can be confusing even for trained specialists causing an information overload exists for many medical doctors. This suggests that content-based image retrieval can be a valuable tool for helping manage these data and access the right information at the right time. This article gives a short state of the art of content-based medical image retrieval followed by a description of the medGIFT project on image retrieval with its main components. Then, several challenges are used to illustrate areas where much more work is currently needed to advance biomedical image retrieval. This shows that we have now progresses beyond the phase, where medical doctors transfer a database to computer scientists to only evaluate their algorithms. We conclude that visual information retrieval can have a real impact in the medical field if the techniques can adapt to this rapidly changing field and get integrated into the workflow in radiology and other medical fields.

Content--based image retrieval in the medical domain is an extremely hot topic in medical imaging as it promises to help better managing the large amount of medical images being produced. Applications are mainly expected in the field of medical teaching files and for research projects, where performance issues and speed are less critical than in the field of diagnostic aid. Final goal with most impact will be the use as a diagnostic aid in a real--world clinical setting.Other applications of image retrieval and image classification can be the automatic annotation of images with basic concepts or the control of DICOM header information.ImageCLEF is part of the Cross Language Evaluation Forum (CLEF). Since 2004, a medical image retrieval task has been added. Goal is to create databases of a realistic and useful size and also query topics that are based on real--world needs in the medical domain but still correspond to the limited capabilities of purely visual retrieval at the moment. Goal is to direct the research onto real applications and towards real clinical problems to give researchers who are not directly linked to medical facilities a possibility to work on the interesting problem of medical image retrieval based on real data sets and problems. The missing link between computer science research departments and clinical routine is one of the biggest problems that becomes evident when reading much of the current literature on medical image retrieval. Most databases are extremely small, the treated problems often far from clinical reality, and there is no integration of the prototypes into a hospital infrastructure. Only few retrieval articles specifically mention problems related to the DICOM format (Digital Imaging and Communications in Medicine) and the sheer amount of data that needs to be treated in an image archive ( > 30.000 images per day in the Geneva radiology).This article develops the various axes that can be taken into account for medical image retrieval system evaluation. First, the axes are developed based on current challenges and experiences from ImageCLEF. Then, the resources developed for ImageCLEF are listed and finally, the application of the axes is explained to show the bases of the ImageCLEFmed evaluation campaign. This article will only concentrate on the medical retrieval tasks, the non-medical tasks will only shortly be mentioned.

BACKGROUND: In the context of the BioCreative competition, where training data were very sparse, we investigated two complementary tasks: 1) given a Swiss-Prot triplet, containing a protein, a GO (Gene Ontology) term and a relevant article, extraction of a short passage that justifies the GO category assignment; 2) given a Swiss-Prot pair, containing a protein and a relevant article, automatic assignment of a set of categories. METHODS: Sentence is the basic retrieval unit. Our classifier computes a distance between each sentence and the GO category provided with the Swiss-Prot entry. The Text Categorizer computes a distance between each GO term and the text of the article. Evaluations are reported both based on annotator judgements as established by the competition and based on mean average precision measures computed using a curated sample of Swiss-Prot. RESULTS: Our system achieved the best recall and precision combination both for passage retrieval and text categorization as evaluated by official evaluators. However, text categorization results were far below those in other data-poor text categorization experiments The top proposed term is relevant in less that 20% of cases, while categorization with other biomedical controlled vocabulary, such as the Medical Subject Headings, we achieved more than 90% precision. We also observe that the scoring methods used in our experiments, based on the retrieval status value of our engines, exhibits effective confidence estimation capabilities. CONCLUSION: From a comparative perspective, the combination of retrieval and natural language processing methods we designed, achieved very competitive performances. Largely data-independent, our systems were no less effective that data-intensive approaches. These results suggests that the overall strategy could benefit a large class of information extraction tasks, especially when training data are missing. However, from a user perspective, results were disappointing. Further investigations are needed to design applicable end-user text mining tools for biologists.

OBJECTIVE: Authors evaluated whether displaying context sensitive links to infrequently accessed educational materials and patient information via the user interface of an inpatient computerized care provider order entry (CPOE) system would affect access rates to the materials. DESIGN: The CPOE of Vanderbilt University Hospital (VUH) included "baseline" clinical decision support advice for safety and quality. Authors augmented this with seven new primarily educational decision support features. A prospective, randomized, controlled trial compared clinicians' utilization rates for the new materials via two interfaces. Control subjects could access study-related decision support from a menu in the standard CPOE interface. Intervention subjects received active notification when study-related decision support was available through context sensitive, visibly highlighted, selectable hyperlinks. MEASUREMENTS: Rates of opportunities to access and utilization of study-related decision support materials from April 1999 through March 2000 on seven VUH Internal Medicine wards. RESULTS: During 4,466 intervention subject-days, there were 240,504 (53.9/subject-day) opportunities for study-related decision support, while during 3,397 control subject-days, there were 178,235 (52.5/subject-day) opportunities for such decision support, respectively (p = 0.11). Individual intervention subjects accessed the decision support features at least once on 3.8% of subject-days logged on (278 responses); controls accessed it at least once on 0.6% of subject-days (18 responses), with a response rate ratio adjusted for decision support frequency of 9.17 (95% confidence interval 4.6-18, p < 0.0005). On average, intervention subjects accessed study-related decision support materials once every 16 days individually and once every 1.26 days in aggregate. CONCLUSION: Highlighting availability of context-sensitive educational materials and patient information through visible hyperlinks significantly increased utilization rates for study-related decision support when compared to "standard" VUH CPOE methods, although absolute response rates were low.

Care providers' adoption of computer-based health-related documentation ("note capture") tools has been limited, even though such tools have the potential to facilitate information gathering and to promote efficiency of clinical charting. The authors have developed and deployed a computerized note-capture tool that has been made available to end users through a care provider order entry (CPOE) system already in wide use at Vanderbilt. Overall note-capture tool usage between January 1, 1999, and December 31, 2001, increased substantially, both in the number of users and in their frequency of use. This case report is provided as an example of how an existing care provider order entry environment can facilitate clinical end-user adoption of a computer-assisted documentation tool-a concept that may seem counterintuitive to some.

We report on the comparison of different strategies for correcting spelling errors resulting in non-existent words. Unlike interactive spelling checkers, where usually only the left context is available, the system we developed takes advantage of the entire context surrounding misspelling. Moreover, unlike traditional systems, based exclusively on a string-to-string edit distance and a word language model, we explore the use of the part-of-speech for selecting candidates. In conclusion, we show that spelling correction improves by extending the context. The best results are obtained when combining a part-of-speech filter with a word language model, and using both the left and right adjacent contexts.

Introduction of a nontrivial clinical information system (CIS) generates major cultural challenges.1 The most obvious relate to the need for health care workers to acquire new skills and do their work in new ways. More subtle challenges relate to alteration of roles and workflow throughout an institution. Implementation of a CIS often undoes myriad “fixes” that were applied over time to make the predecessor manual “nonsystem” processes work better. In this issue of the Journal, Murray et al.2 document shifts in pharmacists' functions caused by the implementation of a computer-based outpatient physician-to-pharmacy prescription system. They show that, for the institution involved, there was no significant improvement in the mechanical efficiency of the pharmacy workflow. Pharmacists spent as much time clarifying the orders that came to them electronically as they previously did entering prescriptions directly. This finding makes clear that the CIS was not designed to shift data entry to the care provider from the clerks and ancillaries. The motivation for implementing an integrated, cross-functional CIS was, and should be, to improve the quality and overall efficiency of patient care. Such systems can be successful when the institution works “smarter,” though not necessarily “faster.” The key goals for CIS include capture of information, preferably once from the most reliable source; delivery of “just-in-time” decision support; and the augmentation and refinement—not the translation—of information as it moves from one area of institutional or clinical expertise to another. Proper design of a computer-based physician prescription-writing system requires clarification of distinct roles for each person who participates in the process. The physician selects the best therapy for the patient and records clinically significant information including the patient ID; the substance(s) to be administered; the dosage, route, and duration; a sense of how urgently the medication must be started; and, optionally, diagnoses or reasons for treatment. Next, the pharmacist augments and refines this clinical information with specific dispensing information including the formulation of the substance (e.g., manufacturer and tablet size), administration times (e.g., the “routine” time for a given inpatient ward or shift), and clerical instructions. Nurses or patients who administer the medication then document the times of administration and report any adverse effects or skipped doses. It is the shared responsibility of the persons selecting the therapy, dispensing it, and administering it to determine whether there are potentially significant drug interactions or allergies that merit consideration of an alternative therapy. Each step represents a different perspective, exemplified by a different set of constraints and business rules. Knowledge bases owned by any participant in the process should be shared in order to move decision support closer to its logical point of use. For example, most hospital pharmacies use, maintain, and customize computer-based systems to check the dosage of drugs and to assist in the detection of drug allergies and drug-drug interactions. The pharmacists have the expertise necessary to maintain this knowledge, but if its use can be moved to the decision-maker at the time of entry of orders by physicians, errors can be prevented before decisions are made final, before they reach the pharmacy. Such strategies can improve mechanical efficiency as well as quality. Similarly, care providers should be responsible for maintaining the list of allergies, as they obtain this information directly from their patients. This list should be available to pharmacists to assist in their review of the treatment plan. In this way participants each do their part to ensure that the other members of the treatment team have the best information on which to base decisions. Realistically, however, both technical and cultural challenges impose compromises during actual implementation. The profound alteration of work flow makes it difficult to fully specify a priori the overall data model or to formalize and externalize business rules. Iterative refinements are necessary. Often, free-text entry is allowed despite its technical inconvenience, in order to provide the expressiveness and ease-of-use required to gain acceptance from clinical users. Such compromises, which are, in essence, temporary, may contribute to the lack of mechanical efficiency observed in the study by Murray et al. When designing, implementing, and evaluating a complex CIS, it is important to acknowledge the necessity of cultural changes, some of which can be anticipated and some of which cannot. Prototyping systems with end-user input, analyzing early and intensely the impact of a new “live” system on people and work flow (informally or more formally, as was done by Murray et al.), and evolving the system rapidly after initial implementation to continually improve work flow and end-user satisfaction are essential components of “clinical informatics.” It is often necessary to implement temporary placeholder algorithms that provide function without compromising the ability to evolve toward a reference architecture. It is crucial to remember that the primary goal of a CIS is to improve the delivery of patient care. The article by Shojania et al. in this issue3 documents the ability of an order entry system to improve compliance with vancomycin prescription guidelines. That work, along with the work of Murray et al., provides an example of why institutions should implement systems that improve clinical decisions at the point of care.

The performance of a pair of multicrystal, high-resolution, bismuth germanate (BGO) block detectors for positron emission tomography (PET) has been investigated. Utilizing the detectors at a separation of 100 mm, the spatial resolutions and count rate response of the block have been measured. These measurements indicate maximum spatial resolutions of 3.6 and 4.5 mm (FWHM) in the two axes of the block and a maximum coincidence rate of 3400 cps. The system has been used to observe the regional kinetics of positron-emitting radionuclides in the rat brain and the human radial artery. Designs for a small diameter, no-septa tomograph incorporating these detectors have been considered. Simulations demonstrate the possibilities of following tracer uptake within the rat brain and radial artery. Three-dimensional tomographic data-sets of a 20-mm uniform cylinder, obtained from rotating the two blocks, indicate good uniformity in the field of view (FOV) of 25 mm.