Infections in Diabetic Foot Ulcers (DFUs) can cause severe complications, including tissue death and limb amputation, highlighting the need for accurate, timely diagnosis. Previous machine learning methods have focused on identifying infections by analyzing wound images alone, without utilizing additional metadata such as medical notes. In this study, we aim to improve infection classification by introducing Synthetic Caption Augmented Retrieval for Wound Infection Detection (SCARWID) , a novel multimodal deep learning framework that leverages synthetic textual descriptions to augment DFU images. SCARWID consists of two components: (1) Wound-BLIP, a Vision-Language Model (VLM) fine-tuned on GPT-4o-generated descriptions to synthesize consistent captions from images; and (2) an Image–Text Fusion module that uses cross-attention to extract cross-modal embeddings from an image and its corresponding Wound-BLIP caption. Infection status is determined by retrieving the top- \( k \) similar items from a labeled support set. To enhance the diversity of training data, we utilized a latent diffusion model to generate additional wound images. As a result, SCARWID outperformed state-of-the-art models, achieving average accuracy, sensitivity, and specificity of 0.814, 0.845, and 0.783, respectively, for wound infection classification.

Automatic assessment of wounds using smartphones can facilitate the tracking of healing progress outside the clinic. Prior work proposed SS-PMG-EfficientNets, a state-of-the-art, multi-scale, convolutional neural network architecture for accurate wound assessment. However, the model was previously only trained and evaluated on high-quality wound images captured in controlled environments. This study systematically evaluated the effects of low-quality images on SS-PMG-EfficientNets, finding that the model's performance and consistency are reduced when input wound images are blurry or captured in adverse lighting (too bright or too dark), two factors previously identified as the most prevalent issues in crowdsourced images. Specifically, SS-PMG-EfficientNet was first trained on three high-quality wound datasets containing images labeled with Photographic Wound Assessment Tool (PWAT) wound healing scores, stages of pressure injury wounds, and wound types, respectively. The model was then evaluated on a test wound dataset that was augmented with real-world image luminance, blur, and image compression effects and found that wound assessment accuracy decreased by up to 27%. To improve model performance, we propose a method for instant image quality assessment on the smartphone, enabling low-quality wound images to be rejected immediately after image capture so that the nurse or patient can be prompted to recapture the image. Our evaluation of the proposed image quality assessment and rejection method showed that it increases average accuracy from 73 % to 87%.

Over one billion smartphones have now been shipped worldwide. These mobile devices feature multi-core CPUs and GPUs, megapixel cameras and an array of sensors. Smartphone sensors can now be processed to diagnose a wide variety of medical conditions including cough detection, irregular heartbeat detection, and lung function analysis. The ability to diagnose ailments in the convenience of patients' homes on smartphones that they already possess could lead to early detection, which could ultimately reduce healthcare costs. This paper reviews state-of-the-art examples, examines the technical issues involved in the use of the smartphone as a medical device, and outlines potential benefits and challenges. A case study of an Android smartphone app for wound detection and healing progress analysis by diabetes patients is also presented.

A key touted benefit of enterprise systems (ES) is organizational integration of both business processes and data, which is expected to reduce processing time and increase control over operations. In our 3-year longitudinal case study of a phased ES implementation, we employed a grounded theory methodology to discover organizational effects of ES. As we coded and analyzed our field data, we observed many integration effects. Further analysis revealed underlying dimensions of ES-enabled integration. ES-enabled integration varied depending on the relationship between the integrated business units (similar plants, stages in a business process, or dissimilar functional areas) and on whether processes or data were integrated. Turning to the literature, we realized that Thompson's three types of interdependence, pooled, sequential, and reciprocal, captured the business relationships revealed in our data. Thus, we describe the salient characteristics of ES-enabled integration using Thompson's interdependence types applied to process and data integration. We also identify dimensions of differentiation between business units that contribute to integration problems. Viewing our field data through the lens of these salient characteristics and dimensions of differentiation provided theoretical explanations for observed integration problems. These findings also help managers understand and anticipate ES-enabled integration opportunities and problems.

Despite the established theory and the history of the practical use of integrity rules, data quality problems, which should be solvable using data integrity rules, persist in organizations. One effective mechanism to solve this problem is to embed data integrity in a continuous data quality improvement process. The result is an iterative data quality improvement process as data integrity rules are defined, violations of these rules are measured and analyzed, and then the rules are redefined to reflect the dynamic and global context of business process changes. Using action research, we study a global manufacturing company that applied these ideas for improving data quality as it built a global data warehouse. This research merges data integrity theory with management theories about quality improvement using a data quality lens, and it demonstrates the usefulness of the combined theory for data quality improvement.

Groupware technologies have become an important part of the business computing infrastructure in many organizations, but many groupware applications, especially those requiring significant collaboration and cooperation among users, are still not adequately used. While the successful implementation of groupware depends on many different factors, achieving a ‘critical mass’ of users has been recognised as the key for groupware acceptance. By extending the technology acceptance model (TAM), this paper advances a groupware acceptance model that incorporates perceived critical mass as an independent variable for predicting groupware acceptance. The model is empirically evaluated using survey data collected from 385 students responding about their perception of Lotus Domino Discussion Databases. The results reveal that perceived critical mass had the largest total effect (direct and indirect) on intention to use groupware. Other relationships postulated in the model were also found to be significant. These findings corroborate the belief that it is essential to create a critical mass of users in the early stages of groupware implementation. A discussion of the implications of these findings for managers and researchers is also presented.

Software tools to support programmers and maintainers have been touted as potential solutions to the software development and maintenance crisis. Use of these tools is predicted to increase programmer productivity while simultaneously increasing the quality of the resulting software. Unfortunately, programmer use of these tools is lower than expected. We investigate maintenance programmers' choices about using software tools for their maintenance tasks using task–technology fit and fitness-for-use models. Our results indicate that the fit between maintenance tool functionality and the needs of maintenance tasks is associated with tool use. Two related factors, however, have stronger effects. First, the results are stronger for intention to use than for actual use. Specifically, while higher fit between tool and task is highly associated with intention to use, this intention may not lead to actual use. Second, maintainers' control over their environment affects usage. The more control maintenance programmers have over their resources and opportunities to use tools, the more likely they are to choose to use them. These results should concern maintenance managers who have acquired or are acquiring tools to increase productivity and quality, but are not realizing the benefits of their technology investments. © 1998 John Wiley & Sons, Ltd.

Exceptions, situations that cannot be correctly processed by computer systems, occur frequently in computer-based information processes. Five perspectives on exceptions provide insights into why exceptions occur and how they might be eliminated or more efficiently handled. We investigate these perspectives using an in-depth study of an operating information process that has frequent exceptions. Our results support the use of a total quality management (TQM) approach of eliminating exceptions for some exceptions, in particular, those caused by computer systems that are poor matches to organizational processes. However, some exceptions are explained better by a political system perspective of conflicting goals between subunits. For these exceptions and several other types, designing an integrated human-computer process will provide better performance than will eliminating exceptions and moving toward an entirely automated process.