Showing papers in "Journal of Medical Systems in 2022"

Health Equity Tourism: Ravaging the Justice Landscape

Abstract: As the long-standing and ubiquitous racial inequities of the United States reached national attention, the public health community has witnessed the rise of "health equity tourism". This phenomenon is the process of previously unengaged investigators pivoting into health equity research without developing the necessary scientific expertise for high-quality work. In this essay, we define the phenomenon and provide an explanation of the antecedent conditions that facilitated its development. We also describe the consequences of health equity tourism - namely, recapitulating systems of inequity within the academy and the dilution of a landscape carefully curated by scholars who have demonstrated sustained commitments to equity research as a primary scientific discipline and praxis. Lastly, we provide a set of principles that can guide novice equity researchers to becoming community members rather than mere tourists of health equity.

Monkeypox Virus Detection Using Pre-trained Deep Learning-based Approaches

Abstract: Monkeypox virus is emerging slowly with the decline of COVID-19 virus infections around the world. People are afraid of it, thinking that it would appear as a pandemic like COVID-19. As such, it is crucial to detect them earlier before widespread community transmission. AI-based detection could help identify them at the early stage. In this paper, we aim to compare 13 different pre-trained deep learning (DL) models for the Monkeypox virus detection. For this, we initially fine-tune them with the addition of universal custom layers for all of them and analyse the results using four well-established measures: Precision, Recall, F1-score, and Accuracy. After the identification of the best-performing DL models, we ensemble them to improve the overall performance using a majority voting over the probabilistic outputs obtained from them. We perform our experiments on a publicly available dataset, which results in average Precision, Recall, F1-score, and Accuracy of 85.44\%, 85.47\%, 85.40\%, and 87.13\%, respectively with the help of our proposed ensemble approach. These encouraging results, which outperform the state-of-the-art methods, suggest that the proposed approach is applicable to health practitioners for mass screening.

Human Monkeypox Classification from Skin Lesion Images with Deep Pre-trained Network using Mobile Application

Abstract: Recently, human monkeypox outbreaks have been reported in many countries. According to the reports and studies, quick determination and isolation of infected people are essential to reduce the spread rate. This study presents an Android mobile application that uses deep learning to assist this situation. The application has been developed with Android Studio using Java programming language and Android SDK 12. Video images gathered through the mobile device’s camera are dispatched to a deep convolutional neural network that runs on the same device. Camera2 API of the Android platform has been used for camera access and operations. The network then classifies images as positive or negative for monkeypox detection. The network’s training has been carried out using skin lesion images of monkeypox-infected people and other skin lesion images. For this purpose, a publicly available dataset and a deep transfer learning approach have been used. All training and testing steps have been applied on Matlab using different pre-trained networks. Then, the network that has the best accuracy has been recreated and trained using TensorFlow. The TensorFlow model has been adapted to mobile devices by converting to the TensorFlow Lite model. The TensorFlow Lite model has been then embedded into the mobile application together with the TensorFlow Lite library for monkeypox detection. The application has been run on three devices successfully. During the run-time, the inference times have been gathered. 197 ms, 91 ms, and 138 ms average inference times have been observed. The presented system allows people with body lesions to quickly make a preliminary diagnosis. Thus, monkeypox-infected people can be encouraged to act rapidly to see an expert for a definitive diagnosis. According to the test results, the system can classify the images with 91.11% accuracy. In addition, the proposed mobile application can be trained for the preliminary diagnosis of other skin diseases.

Trustworthy Augmented Intelligence in Health Care

Abstract: Augmented Intelligence (AI) systems have the power to transform health care and bring us closer to the quadruple aim: enhancing patient experience, improving population health, reducing costs, and improving the work life of health care providers. Earning physicians' trust is critical for accelerating adoption of AI into patient care. As technology evolves, the medical community will need to develop standards for these innovative technologies and re-visit current regulatory systems that physicians and patients rely on to ensure that health care AI is responsible, evidence-based, free from bias, and designed and deployed to promote equity. To develop actionable guidance for trustworthy AI in health care, the AMA reviewed literature on the challenges health care AI poses and reflected on existing guidance as a starting point for addressing those challenges (including models for regulating the introduction of innovative technologies into clinical care).

Domain Shifts in Machine Learning Based Covid-19 Diagnosis From Blood Tests

Abstract: Many previous studies claim to have developed machine learning models that diagnose COVID-19 from blood tests. However, we hypothesize that changes in the underlying distribution of the data, so called domain shifts, affect the predictive performance and reliability and are a reason for the failure of such machine learning models in clinical application. Domain shifts can be caused, e.g., by changes in the disease prevalence (spreading or tested population), by refined RT-PCR testing procedures (way of taking samples, laboratory procedures), or by virus mutations. Therefore, machine learning models for diagnosing COVID-19 or other diseases may not be reliable and degrade in performance over time. We investigate whether domain shifts are present in COVID-19 datasets and how they affect machine learning methods. We further set out to estimate the mortality risk based on routinely acquired blood tests in a hospital setting throughout pandemics and under domain shifts. We reveal domain shifts by evaluating the models on a large-scale dataset with different assessment strategies, such as temporal validation. We present the novel finding that domain shifts strongly affect machine learning models for COVID-19 diagnosis and deteriorate their predictive performance and credibility. Therefore, frequent re-training and re-assessment are indispensable for robust models enabling clinical utility.

Advances in Deep Learning for Tuberculosis Screening using Chest X-rays: The Last 5 Years Review

Abstract: There has been an explosive growth in research over the last decade exploring machine learning techniques for analyzing chest X-ray (CXR) images for screening cardiopulmonary abnormalities. In particular, we have observed a strong interest in screening for tuberculosis (TB). This interest has coincided with the spectacular advances in deep learning (DL) that is primarily based on convolutional neural networks (CNNs). These advances have resulted in significant research contributions in DL techniques for TB screening using CXR images. We review the research studies published over the last five years (2016-2021). We identify data collections, methodical contributions, and highlight promising methods and challenges. Further, we discuss and compare studies and identify those that offer extension beyond binary decisions for TB, such as region-of-interest localization. In total, we systematically review 54 peer-reviewed research articles and perform meta-analysis.

The Influence of Audible Alarm Loudness and Type on Clinical Multitasking

Abstract: In high-consequence industries such as health care, auditory alarms are an important aspect of an informatics system that monitors patients and alerts providers attending to multiple concurrent tasks. Alarms levels are unnecessarily high and alarm signals are uninformative. In a laboratory-based task setting, we studied 25 anesthesiology residents’ responses to auditory alarms in a multitasking paradigm comprised of three tasks: patient monitoring, speech perception/intelligibility, and visual vigilance. These tasks were in the presence of background noise plus/minus music, which served as an attention-diverting stimulus. Alarms signified clinical decompensation and were either conventional alarms or a novel informative auditory icon alarm. Both alarms were presented at four different levels. Task performance (accuracy and response times) were analyzed using logistic and linear mixed-effects regression. Salient findings were 1), the icon alarm had similar performance to the conventional alarm at a +2 dB signal-to-noise-ratio (SNR) (accuracy: OR 1.21 (95% CI 0.88, 1.67), response time: 0.04 s at 2 dB (95% CI: –0.16, 0.24), which is a much lower level than current clinical environments; 2) the icon alarm was associated with 27% greater odds (95% CI: 18%, 37%) of correctly addressing the vigilance task, regardless of alarm SNR, suggesting crossmodal/multisensory multitasking benefits; and 3) compared to the conventional alarm, the icon alarm was associated with an absolute improvement in speech perception of 4% in the presence of an attention-diverting auditory stimulus (p = 0.031). These findings suggest that auditory icons can provide multitasking benefits in cognitively demanding clinical environments.

Online Mindfulness Experience for Emotional Support to Healthcare staff in times of Covid-19

Abstract: During the first confinement in Spain, between the months of March to June 2020, Information and Communication Technologies strategies were implemented in order to support health workers in the Wellbeing of Mental Health. Faced with so much uncertainty about the pandemic, an Online Mindfulness course. The objective of the course was to support healthcare professionals in Castilla y León in managing stress, anxiety and other emotional disturbances generated by coping with a situation as uncertain and unexpected as a pandemic, in order to manage emotions and thoughts that can lead to suicidal ideation. The motivations for the demand, reasons or motivations in which the health professionals of Castilla y León decided to participate in the mindfulness course in the first wave of Covid-19 in Spain are described. The descriptive and inferential statistical analysis of the customer satisfaction survey applied at the end of the mindfulness course, to the health professionals who participated in a satisfaction survey (CSQ-8: Client Satisfaction Questionnaire). Professional were asked to complete a survey based on (CSQ-8: Client Satisfaction Questionnaire) whose Cronbach's alpha = 0.917 is why the instrument used with N = 130 participants has high reliability. The 66% answered with a highly satisfied that they would return to the mindfulness online course. The 93% of the people who answered the satisfaction survey were women, of which they are professionals in the nursing area, with a participation of around 62%. In relation to the online system used in the Mindfulness intervention, 74% expressed that they fully agreed that it has been easy to use the online system for the mindfulness intervention. Health Professionals responded with 58% high satisfaction and 36% satisfaction, making a total of 94% on the help received in the online mindfulness courses to solve their problems. There is no difference between the age groups of the professionals who have preferred the Mindfulness online course (p = 0.672).

Development and Validation of Digital Health Technology Literacy Assessment Questionnaire

Abstract: In clinical practice, assessing digital health literacy is important to identify patients who may encounter difficulties adapting to digital health using digital technology and service. We developed the Digital Health Technology Literacy Assessment Questionnaire (DHTL-AQ) to assess the ability to use digital health technology, services, and data. The DHTL-AQ was developed in three phases. In the first phase, the conceptual framework and domains and items were generated from a systematic literature review using relevant theory and surveys. In the second phase, a cross-sectional survey with 590 adults age ≥ 18 years was conducted at an academic hospital in Seoul, Korea in January and February 2020 to test face validity of the items. Then, psychometric validation was conducted to determine the final items and cut-off scores of the DHTL-AQ. The eHealth literacy scale, the Newest Vital Sign, and 10 mobile app task ability assessments were examined to test validity. The final DHTL-AQ includes 34 items in two domains (digital functional and digital critical literacy) and 4 categories (Information and Communications Technology terms, Information and Communications Technology icons, use of an app, evaluating reliability and relevance of health information). The DHTL-AQ had excellent internal consistency (overall Cronbach's α = 0.95; 0.87-0.94 for subtotals) and acceptable model fit (CFI = 0.821, TLI = 0.807, SRMR = 0.065, RMSEA = 0.090). The DHTL-AQ was highly correlated with task ability assessment (r = 0.7591), and moderately correlated with the eHealth literacy scale (r = 0.5265) and the Newest Vital Sign (r = 0.5929). The DHTL-AQ is a reliable and valid instrument to measure digital health technology literacy.

Clinical Explainability Failure (CEF) & Explainability Failure Ratio (EFR) – Changing the Way We Validate Classification Algorithms

Abstract: Adoption of Artificial Intelligence (AI) algorithms into the clinical realm will depend on their inherent trustworthiness, which is built not only by robust validation studies but is also deeply linked to the explainability and interpretability of the algorithms. Most validation studies for medical imaging AI report the performance of algorithms on study-level labels and lay little emphasis on measuring the accuracy of explanations generated by these algorithms in the form of heat maps or bounding boxes, especially in true positive cases. We propose a new metric - Explainability Failure Ratio (EFR) - derived from Clinical Explainability Failure (CEF) to address this gap in AI evaluation. We define an Explainability Failure as a case where the classification generated by an AI algorithm matches with study-level ground truth but the explanation output generated by the algorithm is inadequate to explain the algorithm's output. We measured EFR for two algorithms that automatically detect consolidation on chest X-rays to determine the applicability of the metric and observed a lower EFR for the model that had lower sensitivity for identifying consolidation on chest X-rays, implying that the trustworthiness of a model should be determined not only by routine statistical metrics but also by novel 'clinically-oriented' models.

Improving the Classification Performance of Esophageal Disease on Small Dataset by Semi-supervised Efficient Contrastive Learning.

Abstract: The classification of esophageal disease based on gastroscopic images is important in the clinical treatment, and is also helpful in providing patients with follow-up treatment plans and preventing lesion deterioration. In recent years, deep learning has achieved many satisfactory results in gastroscopic image classification tasks. However, most of them need a training set that consists of large numbers of images labeled by experienced experts. To reduce the image annotation burdens and improve the classification ability on small labeled gastroscopic image datasets, this study proposed a novel semi-supervised efficient contrastive learning (SSECL) classification method for esophageal disease. First, an efficient contrastive pair generation (ECPG) module was proposed to generate efficient contrastive pairs (ECPs), which took advantage of the high similarity features of images from the same lesion. Then, an unsupervised visual feature representation containing the general feature of esophageal gastroscopic images is learned by unsupervised efficient contrastive learning (UECL). At last, the feature representation will be transferred to the down-stream esophageal disease classification task. The experimental results have demonstrated that the classification accuracy of SSECL is 92.57%, which is better than that of the other state-of-the-art semi-supervised methods and is also higher than the classification method based on transfer learning (TL) by 2.28%. Thus, SSECL has solved the challenging problem of improving the classification result on small gastroscopic image dataset by fully utilizing the unlabeled gastroscopic images and the high similarity information among images from the same lesion. It also brings new insights into medical image classification tasks.

Multicenter Study on COVID-19 Lung Computed Tomography Segmentation with varying Glass Ground Opacities using Unseen Deep Learning Artificial Intelligence Paradigms: COVLIAS 1.0 Validation

Abstract: Variations in COVID-19 lesions such as glass ground opacities (GGO), consolidations, and crazy paving can compromise the ability of solo-deep learning (SDL) or hybrid-deep learning (HDL) artificial intelligence (AI) models in predicting automated COVID-19 lung segmentation in Computed Tomography (CT) from unseen data leading to poor clinical manifestations. As the first study of its kind, "COVLIAS 1.0-Unseen" proves two hypotheses, (i) contrast adjustment is vital for AI, and (ii) HDL is superior to SDL. In a multicenter study, 10,000 CT slices were collected from 72 Italian (ITA) patients with low-GGO, and 80 Croatian (CRO) patients with high-GGO. Hounsfield Units (HU) were automatically adjusted to train the AI models and predict from test data, leading to four combinations-two Unseen sets: (i) train-CRO:test-ITA, (ii) train-ITA:test-CRO, and two Seen sets: (iii) train-CRO:test-CRO, (iv) train-ITA:test-ITA. COVILAS used three SDL models: PSPNet, SegNet, UNet and six HDL models: VGG-PSPNet, VGG-SegNet, VGG-UNet, ResNet-PSPNet, ResNet-SegNet, and ResNet-UNet. Two trained, blinded senior radiologists conducted ground truth annotations. Five types of performance metrics were used to validate COVLIAS 1.0-Unseen which was further benchmarked against MedSeg, an open-source web-based system. After HU adjustment for DS and JI, HDL (Unseen AI) > SDL (Unseen AI) by 4% and 5%, respectively. For CC, HDL (Unseen AI) > SDL (Unseen AI) by 6%. The COVLIAS-MedSeg difference was < 5%, meeting regulatory guidelines.Unseen AI was successfully demonstrated using automated HU adjustment. HDL was found to be superior to SDL.

All Patient Refined-Diagnosis Related Groups’ (APR-DRGs) Severity of Illness and Risk of Mortality as predictors of in-hospital mortality

Abstract: The aims of this study were to assess All-Patient Refined Diagnosis-Related Groups' (APR-DRG) Severity of Illness (SOI) and Risk of Mortality (ROM) as predictors of in-hospital mortality, comparing with Charlson Comorbidity Index (CCI) and Elixhauser Comorbidity Index (ECI) scores. We performed a retrospective observational study using mainland Portuguese public hospitalizations of adult patients from 2011 to 2016. Model discrimination (C-statistic/ area under the curve) and goodness-of-fit (R-squared) were calculated. Our results comprised 4,176,142 hospitalizations with 5.9% in-hospital deaths. Compared to the CCI and ECI models, the model considering SOI, age and sex showed a statistically significantly higher discrimination in 49.6% (132 out of 266) of APR-DRGs, while in the model with ROM that happened in 33.5% of APR-DRGs. Between these two models, SOI was the best performer for nearly 20% of APR-DRGs. Some particular APR-DRGs have showed good discrimination (e.g. related to burns, viral meningitis or specific transplants). In conclusion, SOI or ROM, combined with age and sex, perform better than more widely used comorbidity indices. Despite ROM being the only score specifically designed for in-hospital mortality prediction, SOI performed better. These findings can be helpful for hospital or organizational models benchmarking or epidemiological analysis.

A System for Computational Assessment of Hand Hygiene Techniques

Abstract: The World Health Organization (WHO) recommends a six-step hand hygiene technique. Although multiple studies have reported that this technique yields inadequate skin coverage outcomes, they have relied on manual labeling that provided low-resolution estimations of skin coverage outcomes. We have developed a computational system to precisely quantify hand hygiene outcomes and provide high-resolution skin coverage visualizations, thereby improving hygiene techniques. We identified frequently untreated areas located at the dorsal side of the hands around the abductor digiti minimi and the first dorsal interosseous. We also estimated that excluding Steps 3, 6R, and 6L from the six-step hand hygiene technique leads to cumulative coverage loss of less than 1%, indicating the potential redundancy of these steps. Our study demonstrates that the six-step hand hygiene technique could be improved to reduce the untreated areas and remove potentially redundant steps. Furthermore, our system can be used to computationally validate new proposed techniques, and help optimise hand hygiene procedures.

The Impact of Structured and Standardized Documentation on Documentation Quality; a Multicenter, Retrospective Study

Abstract: The reuse of healthcare data for various purposes will become increasingly important in the future. To enable the reuse of clinical data, structured and standardized documentation is conditional. However, the primary purpose of clinical documentation is to support high-quality patient care. Therefore, this study investigated the effect of increased structured and standardized documentation on the quality of notes in the Electronic Health Record. A multicenter, retrospective design was used to assess the difference in note quality between 144 unstructured and 144 structured notes. Independent reviewers measured note quality by scoring the notes with the Qnote instrument. This instrument rates all note elements independently using and results in a grand mean score on a 0-100 scale. The mean quality score for unstructured notes was 64.35 (95% CI 61.30-67.35). Structured and standardized documentation improved the Qnote quality score to 77.2 (95% CI 74.18-80.21), a 12.8 point difference (p < 0.001). Furthermore, results showed that structured notes were significantly longer than unstructured notes. Nevertheless, structured notes were more clear and concise. Structured documentation led to a significant increase in note quality. Moreover, considering the benefits of structured data recording in terms of data reuse, implementing structured and standardized documentation into the EHR is recommended.

Improving usability of Electronic Health Records in a UK Mental Health setting: a feasibility study

Abstract: Electronic Health Records (EHRs) can help clinicians to plan, document and deliver care for patients in healthcare services. When used consistently, EHRs can advance patient safety and quality, and reduce clinician's workload. However, usability problems can make it difficult for clinicians to use EHRs effectively, which can negatively impact both healthcare professionals and patients.To improve usability of EHRs within a mental health service in the UK.This was a feasibility study conducted with two mental health teams. A mixed-methods approach was employed. Focus group discussions with clinicians identified existing usability problems in EHRs and changes were made to address these problems. Updated EHR assessment forms were evaluated by comparing the following measures pre and post changes: (1) usability testing to monitor time spent completing and duplicating patient information in EHRs, (2) clinician's experience of using EHRs, and (3) proportion of completed EHR assessment forms.Usability testing with clinicians (n = 3) showed that the time taken to complete EHR assessment forms and time spent duplicating patient information decreased. Clinician's experience of completing EHR assessment forms also significantly improved post changes compared to baseline (n = 71; p < 0.005). There was a significant increase in completion of most EHR forms by both teams after EHR usability improvements (all at p < 0.01).Usability improvements to EHRs can reduce the time taken to complete forms, advance clinician's experience and increase usage of EHRs. It is important to engage healthcare professionals in the usability improvement process of EHRs in mental health services.

Regionalized Health System Review of Automated Dispensing Cabinet Overrides.

Abstract: Automated dispensing cabinet (ADC) overrides are used to access emergent or urgent medications when time delay from computerized provider order entry may result in patient harm. [1] Although necessary, ADC overrides bypass the safety features of order entry and verification which increase the risk of an error occurring and potential patient harm. To protect patient safety, national organizations such as The Joint Commission and Institute for Safe Medication Practices have called for hospitals to review overriding trends and available medications on override. AdventHealth Central Florida Division – South (CFD-S) met the recommendations to track overrides but there was limited understanding of the data. A quality improvement project was necessary to investigate the division’s risk of error and identify interventions to proactively limit patient risk. The initial task of the quality improvement project was to create a standardized ADC override report that could be shared with pharmacy and nursing leaders within AdventHealth CFD-S monthly. As the project progressed, multiple interventions were identified such as standardizing the information reflected in the report, improving education about ADC overrides across multi-disciplinary departments, and critically reviewing the data to identify needed changes within the division. The efforts to share the ADC override metrics across all levels has improved understanding of ADC override goals and intentions of monitoring ADC overrides. This has paved the path to improving ADC override unit culture and identify gaps within the system that allows overrides to occur.

Automatic Cardiac Structure Contouring for Small Datasets with Cascaded Deep Learning Models

Abstract: Cardiac structure contouring is a time consuming and tedious manual activity used for radiotherapeutic dose toxicity planning. We developed an automatic cardiac structure segmentation pipeline for use in low-dose non-contrast planning CT based on deep learning algorithms for small datasets. Fifty CT scans were retrospectively selected and the whole heart, ventricles and atria were contoured. A two stage deep learning pipeline was trained on 41 non contrast planning CTs, tuned with 3 CT scans and validated on 6 CT scans. In the first stage, An InceptionResNetV2 network was used to identify the slices that contained cardiac structures. The second stage consisted of three deep learning models trained on the images containing cardiac structures to segment the structures. The three deep learning models predicted the segmentations/contours on axial, coronal and sagittal images and are combined to create the final prediction. The final accuracy of the pipeline was quantified on 6 volumes by calculating the Dice similarity coefficient (DC), 95% Hausdorff distance (95% HD) and volume ratios between predicted and ground truth volumes. Median DC and 95% HD of 0.96, 0.88, 0.92, 0.80 and 0.82, and 1.86, 2.98, 2.02, 6.16 and 6.46 were achieved for the whole heart, right and left ventricle, and right and left atria respectively. The median differences in volume were -4, -1, + 5, -16 and -20% for the whole heart, right and left ventricle, and right and left atria respectively. The automatic contouring pipeline achieves good results for whole heart and ventricles. Robust automatic contouring with deep learning methods seems viable for local centers with small datasets.

Text Messaging as a Communication Modality to Promote Screening Mammography in Low-income African American Women

Abstract: Though text messages are increasingly used in health promotion, the current understanding of text message-based interventions to increase screening mammography in low-income African American women is limited. This study aimed to assess the feasibility and acceptability of a text message-based intervention to increase screening mammography in low-income African American women.A 15-item, self-administered, paper-based survey on cell phone ownership, text messaging practices and preferences for future breast health information was administered to 120 female patients at an urban family medicine office. Descriptive analyses and demographic correlates of text messaging practices and preferences were examined.The majority of respondents (95%) were cell phone owners of whom 81% reported texting. Prior receipt of a text message from a doctor's office was reported by 51% of cell phone owners. Mammography appointment reminders were the most desired content for future breast health text messages. Age (≥ 70 years old) was found to have a significant negative relationship with text messaging practices and perceptions.The use of text messages to promote mammography was found to be acceptable in this patient population. In addition to age, variables such as the frequency, timing and subject content of text messages also influence their acceptability.

Potential Usefulness of Tracking Head Movement via a Wearable Device for Equilibrium Function Testing at Home

Abstract: Many studies have reported the use of wearable devices to acquire biological data for the diagnosis and treatment of various diseases. Balance dysfunction, however, is difficult to evaluate in real time because the equilibrium function is conventionally examined using a stabilometer installed on the ground. Here, we used a wearable accelerometer that measures head motion to evaluate balance and examined whether it performs comparably to a conventional stabilometer. We constructed a simplified physical head-feet model that simultaneously records "head" motion measured using an attached wearable accelerometer and center-of-gravity motion at the "feet", which is measured using an attached stabilometer. Total trajectory length (r = 0.818, p -false discovery rate [FDR] = 0.004) and outer peripheral area (r = 0.691, p -FDR = 0.026) values measured using the wearable device and stabilometer were significantly positively correlated. Root mean square area values were not significantly correlated with wearable device stabilometry but were comparable. These results indicate that wearable, widely available, non-medical devices may be used to assess balance outside the hospital setting, and new approaches for testing balance function should be considered.

Dicoogle Open Source: The Establishment of a New Paradigm in Medical Imaging

Abstract: The rapid and continuous growth of data volume and its heterogeneity has become one of the most noticeable trends in healthcare, namely in medical imaging. This evolution led to the deployment of specialized information systems supported by the DICOM standard that enables the interoperability of distinct components, including imaging modalities, repositories, and visualization workstations. However, the complexity of these ecosystems leads to challenging learning curves and makes it time-consuming to mock and apply new ideas. Dicoogle is an extensible medical imaging archive server that emerges as a tool to overcome those challenges. Its extensible architecture allows the fast development of new advanced features or extends existent ones. It is currently a fundamental enabling technology in collaborative and telehealthcare environments, including research projects, screening programs, and teleradiology services. The framework is supported by a Learning Pack that includes a description of the web programmatic interface, a software development kit, documentation, and implementation samples. This article gives an in-depth view of the Dicoogle ecosystem, state-of-the-art contributions, and community impact. It starts by presenting an overview of its architectural concept, highlights some of the most representative research backed up by Dicoogle, some remarks obtained from its use in teaching, and worldwide usage statistics of the software. Finally, the positioning of Dicoogle in the medical imaging software field is discussed through comparison with other well-known solutions.

The Impact of Functional Dependence and Related Surgical Complications on Postoperative Mortality

Abstract: Functional dependency is a known determinant of surgical risk. To enhance our understanding of the relationship between dependency and adverse surgical outcomes, we studied how postoperative mortality following a surgical complication was impacted by preoperative functional dependency. We explored a historical cohort of 6,483,387 surgical patients within the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). All patients ≥ 18 years old within the ACS-NSQIP from 2007 to 2017 were included. There were 6,222,611 (96.5%) functionally independent, 176,308 (2.7%) partially dependent, and 47,428 (0.7%) totally dependent patients. Within 30 days postoperatively, 57,652 (0.9%) independent, 15,075 (8.6%) partially dependent, and 10,168 (21.4%) totally dependent patients died. After adjusting for confounders, increasing functional dependency was associated with increased odds of mortality (Partially Dependent OR: 1.72, 99% CI: 1.66 to 1.77; Totally Dependent OR: 2.26, 99% CI: 2.15 to 2.37). Dependency also significantly impacted mortality following a complication; however, independent patients usually experienced much stronger increases in the odds of mortality. There were six complications not associated with increased odds of mortality. Model diagnostics show our model was able to distinguish between patients who did and did not suffer 30-day postoperative mortality nearly 96.7% of the time. Within our cohort, dependent surgical patients had higher rates of comorbidities, complications, and odds of 30-day mortality. Preoperative functional status significantly impacted the level of postoperative mortality following a complication, but independent patients were most affected.