Automation Bias in AI-Decision Support: Results from an Empirical Study.

INTRODUCTION: Automation bias poses a significant challenge to the effectiveness of Clinical Decision Support Systems (CDSS), potentially compromising diagnostic accuracy. Previous research highlights trust, self-confidence, and task difficulty as key determinants. With the increasing availability of AI-enabled CDSS, automation bias attains new attention. This study therefore aims to identify factors influencing automation bias in a diagnostic task. METHODS: A quantitative intervention study with participants from different backgrounds (n = 210) was conducted, employing regression analysis to analyze potential factors. Automation bias was measured as the agreement rate with wrong AI-enabled recommendations. RESULTS AND DISCUSSION: Diagnostic performance, certified wound care training, physician profession, and female gender significantly reduced false agreement rates. Higher perceived benefit of the system was significantly associated with promoting false agreement. Strategies like comprehensive diagnostic training are pivotal in the prevention of automation bias when implementing CDSS. CONCLUSION: Considering factors influencing automation bias when introducing a CDSS is critical to fully leverage the benefits of such a system. This study highlights that non-specialists, who stand to gain the most from CDSS, are also the most susceptible to automation bias, emphasizing the need for specialized training to mitigate this risk and ensure diagnostic accuracy and patient safety.

Automatic Classification of Wound Images Showing Healing Complications: Towards an Optimised Approach for Detecting Maceration.

This study aims to advance the field of digital wound care by developing and evaluating convolutional neural network (CNN) architectures for the automatic classification of maceration, a significant wound healing complication, in 458 annotated wound images. Detection and classification of maceration can improve patient outcomes. Several CNN models were compared and MobileNetV2 emerged as the top-performing model, achieving the highest accuracy despite having fewer parameters. This finding underscores the importance of considering model complexity relative to dataset size. The study also explored the role of image cropping and the use of Grad-CAM visualizations to understand the decision-making process of the CNN. From a medical perspective, results indicate that employing CNNs for classification of maceration may enhance diagnostic accuracy and reduce the clinicians' time and effort.

Adoption and Determinants of Assistive Technologies in the Real World: Results from the VdK Study.

INTRODUCTION: While there is growing evidence of the benefits of assistive technologies little is known about their adoption under real circumstances and prevalence for everyday use. OBJECTIVE: The aim of this analysis therefore was (i) to investigate the adoption rates in the real world and (ii) to identify potential determinants of their adoption by care-dependant persons and family caregivers. METHODS: The present study is a secondary analysis based on the data set of the VdK study on home care arrangements (n=53,678). The analysis of the adoption rates included 22,666 care-dependant persons and caregivers, the identification of potential determinants via binary logistic regressions included 5,275 persons. RESULTS: Emergency call systems and technical (smart) aids reached an adoption rate of 40.4 % (care-dependant persons) and 55.3 % (family caregivers). Fall detectors, orientations aids, nursing apps and monitoring systems were used in less than 5 % of the cases. Care degree and the use of an ambulatory nursing service increased the likelihood of using technical aids. CONCLUSION: It can be concluded that innovative and sophisticated types of assistive technologies are still rather scarcely used for home care arrangements in the real world despite large research efforts in the last twenty years.

Design and Implementation of an ETL-Process to Transfer Wound-Related Data into a Standardized Common Data Model.

For observational studies, which are relevant especially for chronic conditions like chronic wounds, the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) offers a standardized database schema. In this study an ETL process for the transition of wound related data was developed. After understanding the data in general and mapping the relevant codes to concepts available in OMOP, the ETL process was implemented. In a first step, a generic algorithm to convert data to a csv format was implemented in Java. The resulting csv file was then processed within KNIME to be loaded into an OMOP CDM conformant database. During the whole ETL process, HL7 FHIR CodeSystem and ConceptMap resources were used for coding and mapping. First clinical test cases to retrieve data were successfully processed as an example to demonstrate the feasibility and usefulness. They concerned wound size at the first visit and the main issues of patients in the wound quality of life questionnaire (n = 24). In general, the ETL process worked well, yet some challenges arose, like post coordinated SNOMED codes or conditions, which might occur more than once.

Predilection sites of pyoderma gangrenosum: Retrospective study of 170 clearly diagnosed patients.

Pyoderma gangrenosum (PG) is a non-infectious, neutrophilic dermatosis that was difficult to diagnose in clinical practice. Today, the PARACELSUS score is a validated tool for diagnostics. Based on this score, patients with clearly diagnosed PG were examined with regard to predilection sites. In this retrospective study, the data of patients from the University Hospitals of Essen and Erlangen were analysed in whom the diagnosis of PG could be clearly confirmed using the PARACELSUS score. A total of 170 patients, 49 men (29%) and 121 women (71%) with an average age at first manifestation of 55.5 years, could be included in the analysis. The predilection sites were identified as the lower legs in 80.6% of the patients and the extensor sides in 75.2%. Other localisations of PG were the thighs in 14.1%, mammae and abdomen in 10.0% each, back and gluteal in 7.1% each, feet in 5.9%, arms in 4.7%, genital in 3.5% and head in 2.9%. This retrospective study is the first to identify a collective of PG patients with the highest data quality using the PARACELSUS score. It could be shown that PG can basically occur on the entire integument. However, the predilection sites of PG, which have now been reliably identified for the first time, are the lower legs and in particular the extensor sides.

Can Synthetic Images Improve CNN Performance in Wound Image Classification?

For artificial intelligence (AI) based systems to become clinically relevant, they must perform well. Machine Learning (ML) based AI systems require a large amount of labelled training data to achieve this level. In cases of a shortage of such large amounts, Generative Adversarial Networks (GAN) are a standard tool for synthesising artificial training images that can be used to augment the data set. We investigated the quality of synthetic wound images regarding two aspects: (i) improvement of wound-type classification by a Convolutional Neural Network (CNN) and (ii) how realistic such images look to clinical experts (n = 217). Concerning (i), results show a slight classification improvement. However, the connection between classification performance and the size of the artificial data set is still unclear. Regarding (ii), although the GAN could produce highly realistic images, the clinical experts took them for real in only 31% of the cases. It can be concluded that image quality may play a more significant role than data size in improving the CNN-based classification result.

[Homecare arrangements: examining the use of support services in relation to socioeconomic status]. / Häusliche Pflegearrangements: Untersuchung der Inanspruchnahme von Unterstützungsleistungen im Zusammenhang mit dem sozioökonomischen Status.

BACKGROUND: National and international findings indicate that homecare arrangements are influenced by socioeconomic factors. Demographic and social developments justify the assumption of an increasing number of people in need of care from resource-pour groups and therefore a continuously increasing importance of home-based long-term care. OBJECTIVE: It is examined whether income and education of people in need of care and their caregivers are related to the use of support services in homecare arrangements. MATERIAL AND METHODS: Quantitative secondary data analysis of a survey among members of the social association VdK (Sozialverband VdK). The statistical evaluation includes descriptive analyses as well as the investigation of correlations between socioeconomic characteristics and characteristics of the care arrangement. RESULTS: A correlation between income and the use of support services exists regarding so-called 24-hour care, which is more common with high incomes. Other income effects can be seen in the extent to which care services and home help services are used and in measures to adapt the home environment. A higher level of education is related to an increased use of counselling services. The assessment of the care situation is more negative the lower the income. CONCLUSIONS: The results point out that phenomena of social inequality in care exist and that the options for shaping homecare are influenced by socioeconomic factors. The study also shows challenges in dealing with social inequality and provides orientation for further research, which is becoming increasingly important in view of current trends.

What Went Wrong in eMedCare? Formative Evaluation of an IT Project in Primary Care in Two Rural Districts.

INTRODUCTION: The interaction between nurses and physicians in the primary care setting is challenging with regard to structural, process and technical barriers. In order to overcome these barriers, the eMedCare project was launched and a commercial system was implemented. OBJECTIVE: This study aimed at a formative evaluation of the project. The findings should be used retrospectively to understand the failure of the project. METHODS: To this end, two rounds of qualitative interviews with 10 respectively 8 healthcare providers were performed. RESULTS: The interviews revealed a mixed benefit. Difficulties arose because the initial aim to monitor patients shifted towards improving the communication between the providers, partly due to the poor usability of the monitoring system. Additional workload was imposed because the system was not interoperable with the institutional IT systems. CONCLUSION: Projects with an unclear or shifting vision and focus seem to be susceptible to failure. The secure communication applications could have been realised on the intended scale if the national Telematikinfrastruktur had been in place.

Automatic Wound Type Classification with Convolutional Neural Networks.

Chronic wounds are ulcerations of the skin that fail to heal because of an underlying condition such as diabetes mellitus or venous insufficiency. The timely identification of this condition is crucial for healing. However, this identification requires expert knowledge unavailable in some care situations. Here, artificial intelligence technology may support clinicians. In this study, we explore the performance of a deep convolutional neural network to classify diabetic foot and venous leg ulcers using wound images. We trained a convolutional neural network on 863 cropped wound images. Using a hold-out test set with 80 images, the model yielded an F1-score of 0.85 on the cropped and 0.70 on the full images. This study shows promising results. However, the model must be extended in terms of wound images and wound types for application in clinical practice.

An Image Based Object Recognition System for Wound Detection and Classification of Diabetic Foot and Venous Leg Ulcers.

Venous leg ulcers and diabetic foot ulcers are the most common chronic wounds. Their prevalence has been increasing significantly over the last years, consuming scarce care resources. This study aimed to explore the performance of detection and classification algorithms for these types of wounds in images. To this end, algorithms of the YoloV5 family of pre-trained models were applied to 885 images containing at least one of the two wound types. The YoloV5m6 model provided the highest precision (0.942) and a high recall value (0.837). Its mAP_0.5:0.95 was 0.642. While the latter value is comparable to the ones reported in the literature, precision and recall were considerably higher. In conclusion, our results on good wound detection and classification may reveal a path towards (semi-) automated entry of wound information in patient records. To strengthen the trust of clinicians, we are currently incorporating a dashboard where clinicians can check the validity of the predictions against their expertise.

Automatic Classification of Diabetic Foot Ulcer Images - A Transfer-Learning Approach to Detect Wound Maceration.

Diabetic foot ulcer (DFU) is a chronic wound and a common diabetic complication as 2% - 6% of diabetic patients witness the onset thereof. The DFU can lead to severe health threats such as infection and lower leg amputations, Coordination of interdisciplinary wound care requires well-written but time-consuming wound documentation. Artificial intelligence (AI) systems lend themselves to be tested to extract information from wound images, e.g. maceration, to fill the wound documentation. A convolutional neural network was therefore trained on 326 augmented DFU images to distinguish macerated from unmacerated wounds. The system was validated on 108 unaugmented images. The classification system achieved a recall of 0.69 and a precision of 0.67. The overall accuracy was 0.69. The results show that AI systems can classify DFU images for macerations and that those systems could support clinicians with data entry. However, the validation statistics should be further improved for use in real clinical settings. In summary, this paper can contribute to the development of methods to automatic wound documentation.

Expressiveness of an International Semantic Standard for Wound Care: Mapping a Standardized Item Set for Leg Ulcers to the Systematized Nomenclature of Medicine-Clinical Terms.

BACKGROUND: Chronic health conditions are on the rise and are putting high economic pressure on health systems, as they require well-coordinated prevention and treatment. Among chronic conditions, chronic wounds such as cardiovascular leg ulcers have a high prevalence. Their treatment is highly interdisciplinary and regularly spans multiple care settings and organizations; this places particularly high demands on interoperable information exchange that can be achieved using international semantic standards, such as Systematized Nomenclature of Medicine-Clinical Terms (SNOMED CT). OBJECTIVE: This study aims to investigate the expressiveness of SNOMED CT in the domain of wound care, and thereby its clinical usefulness and the potential need for extensions. METHODS: A clinically consented and profession-independent wound care item set, the German National Consensus for the Documentation of Leg Wounds (NKDUC), was mapped onto the precoordinated concepts of the international reference terminology SNOMED CT. Before the mapping took place, the NKDUC was transformed into an information model that served to systematically identify relevant items. The mapping process was carried out in accordance with the ISO/TR 12300 formalism. As a result, the reliability, equivalence, and coverage rate were determined for all NKDUC items and sections. RESULTS: The developed information model revealed 268 items to be mapped. Conducted by 3 health care professionals, the mapping resulted in moderate reliability (κ=0.512). Regarding the two best equivalence categories (symmetrical equivalence of meaning), the coverage rate of SNOMED CT was 67.2% (180/268) overall and 64.3% (108/168) specifically for wounds. The sections general medical condition (55/66, 83%), wound assessment (18/24, 75%), and wound status (37/57, 65%), showed higher coverage rates compared with the sections therapy (45/73, 62%), wound diagnostics (8/14, 57%), and patient demographics (17/34, 50%). CONCLUSIONS: The results yielded acceptable reliability values for the mapping procedure. The overall coverage rate shows that two-thirds of the items could be mapped symmetrically, which is a substantial portion of the source item set. Some wound care sections, such as general medical conditions and wound assessment, were covered better than other sections (wound status, diagnostics, and therapy). These deficiencies can be mitigated either by postcoordination or by the inclusion of new concepts in SNOMED CT. This study contributes to pushing interoperability in the domain of wound care, thereby responding to the high demand for information exchange in this field. Overall, this study adds another puzzle piece to the general knowledge about SNOMED CT in terms of its clinical usefulness and its need for further extensions.

Requirements for Collaborative Decision Support Systems in Wound Care: No Information Continuity Without Management Continuity.

Health IT systems are employed to support continuity of care via information continuity, while management continuity is often neglected. This study aims at investigating issues of management continuity when developing a collaborative decision support system for chronic wounds. Thirty-three experts from a variety of professions and disciplines discussed problems and possible solutions in four workshops. The following topics emerged from the discussion: existing networks involving payers, responsibilities as well as good discharge management. These topics clearly address management continuity and are also relevant for the scenario of inter-professional wound care across different settings.

Cognitive Performance of Users Is Affected by Electronic Handovers Depending on Role, Task and Human Factors.

Patient handovers are cognitively demanding, crucial for information continuity and patient safety, but error prone. This study investigated the effect of an electronic handover tool, i.e. the handoverEHR, on the memory and care planning performance of nurse students (n=32) in a randomised, controlled cross-over design with the factors handover task and handover role. On a descriptive level, handover recipients could improve their memory performance with electronic support, handover givers their performance of writing care plans. Statistically meaningful differences occurred, however, only when the participants were givers. Without handover experience and with low fluency to word problems, givers performed badly in the most demanding of the handover tasks. Final recommendations, however, can only be made after replicating this study in a clinical setting with mixed groups.

Patient Handovers - Cognitively Demanding: Does the Handover EHR Meet This Challenge?

Patient handovers are crucial to warrant continuity of care and patient safety. The handoverEHR is an instrument that allows users to depict the main features of a clinical case in cognitive maps. We were interested whether this tool had an effect on the task load. A cross-over study with 30 nursing students was therefore conducted. Mental demand showed a statistical trend to be lower in cognitive map handovers than in the other handover types.

The Benefits of a Formative Evaluation for Developing a Highly Innovative Software: The Case of the handoverEHR.

Innovations are typically characterised by their relative newness for the user. In order for new eHealth applications to be accepted as innovations more criteria were proposed including "use" and "usability". The handoverEHR is a new approach that allows the user to translate the essentials of a clinical case into a graphical representation, the so-called cognitive map of the patient. This study aimed at testing the software usability. A convenience sample of 23 experienced nurses from different healthcare organisations across the country rated the usability of the handoverEHR after performing typical handover tasks. All usability scales of the IsoMetricsL questionnaire showed positive values (4 "I agree") with the exception of "error tolerance" (3 "neutral statement"). A significant improvement was found in self-descriptiveness as compared to an initial usability testing prior to this study. Different subgroups of users tended to rate the usability of the system differently. This study demonstrated the benefits of formative evaluations in terms of improving the usability of an entirely new approach. It thus helps to transform a novel piece of software towards becoming a real innovation. Our findings also hint at the importance of user characteristics that could affect the usability ratings.

Restricted Versus Unrestricted Search Space: Experience from Mining a Large Japanese Database.

The aim of this study was to investigate whether standard Big Data mining methods lead to clinically useful results. An association analysis was performed using the apriori algorithm to discover associations among co-morbidities of diabetes patients. Selected data were further analyzed by using k-means clustering with age, long-term blood sugar and cholesterol values. The association analysis led to a multitude of trivial rules. Cluster analysis detected clusters of well and badly managed diabetes patients both belonging to different age groups. The study suggests the usage of cluster analysis on a restricted space to come to meaningful results.