Clinical Support System Research Articles

Effect of Didactic Training on Barriers and Biases to Treatment of Opioid Use Disorder: Meeting the Ongoing Needs of Patients with Opioid Use Disorder in the Emergency Department during the COVID-19 Pandemic.

In the wake of COVID-19, morbidity and mortality due to Opioid Use Disorder (OUD) is beginning to emerge as a second wave of deaths of despair. Medication assisted treatment (MAT) for opioid use disorder MAT delivered by Emergency Medicine (EM) providers can decrease mortality due to OUD; however, there are numerous cited barriers to MAT delivery. We examined the impact of MAT training on these barriers among EM residents in an urban, tertiary care facility with a large EM residency. Training included the scripted and standardized content from the Provider Clinical Support System curriculum. Residents completed pre- and post-training surveys on knowledge, barriers, and biases surrounding OUD. We performed Wilcoxon matched-pairs signed-ranks test to detect statistical differences. Of 74 residents, 49 (66%) completed the pre-training survey, and 34 (69%) of these completed the follow-up survey. Residents reported improved preparedness to treat aspects of OUD across all areas queried, reported decreased perception of barriers to providing MAT, and increased comfort prescribing naloxone, counseling patients, prescribing buprenorphine, and treating opioid withdrawal. A didactic training on MAT was associated with residents reporting improved comfort providing buprenorphine and naloxone. As the wake of morbidity and mortality from both COVID and OUD continue to increase, programs should offer dedicated training on MAT.

The Role of Data Analytics in the Assessment of Pathological Speech—A Critical Appraisal

Pathological voice characterization has received increasing attention over the last 20 years. Hundreds of studies have been published showing inventive approaches with very promising findings. Nevertheless, methodological issues might hamper performance assessment trustworthiness. This study reviews some critical aspects regarding data collection and processing, machine learning-oriented methods, and grounding analytical approaches, with a view to embedding developed clinical decision support tools into the diagnosis decision-making process. A set of 26 relevant studies published since 2010 was selected through critical selection criteria and evaluated. The model-driven (MD) or data-driven (DD) character of the selected approaches is deeply examined considering novelty, originality, statistical robustness, trustworthiness, and clinical relevance. It has been found that before 2020 most of the works examined were more aligned with MD approaches, whereas over the last two years a balanced proportion of DD and MD-based studies was found. A total of 15 studies presented MD characters, whereas seven were mainly DD-oriented, and four shared both profiles. Fifteen studies showed exploratory or prospective advanced statistical analysis. Eighteen included some statistical validation to avail claims. Twenty-two reported original work, whereas the remaining four were systematic reviews of others’ work. Clinical relevance and acceptability by voice specialists were found in 14 out of the 26 works commented on. Methodological issues such as detection and classification performance, training and generalization capability, explainability, preservation of semantic load, clinical acceptance, robustness, and development expenses have been identified as major issues in applying machine learning to clinical support systems. Other important aspects to be taken into consideration are trustworthiness, gender-balance issues, and statistical relevance.

Implementation of a clinical trial recruitment support system based on fast healthcare interoperability resources (FHIR) in a cardiology department

Abstract Background Clinical Trial Recruitment Support Systems can booster patient inclusion of clinical trials by automatically analyzing eligibility criteria based on electronic health records. However, missing interoperability has hindered introduction of those systems on a broader scale. Purpose Our aim was to develop a recruitment support system based on FHIR R4 and evaluate its usage and features in a cardiology department. Methods/Implementation Clinical conditions, anamnesis, examinations, allergies, medication, laboratory data and echocardiography results were imported as FHIR resources. Trial study nurses and physicians were enabled to add new and edit trial information and input inclusion and exclusion criteria using a web-browser user interface in the hospital intranet. All information were recorded on the server side as the FHIR resources “ResearchStudy” and “Group”. Eligibility criteria linked by the logical operation “OR” were represented by using multiple FHIR Group resources for enrollment. On the client side, eligibility criteria were transformed to a tree-like structure (see Figure 1). Upon user demand, all hospitalized and ambulatory patients in the cardiology department were instantly screened for trial eligibility using the FHIR eligibility criteria on the existing patients' FHIR resources. Furthermore, study personal was able to manually edit trial status (i.e. ineligible, on-study, ...) of patients, which was implemented using the FHIR resource “ResearchSubject”. Results This implementation of a CTRSS based on FHIR R4 was evaluated in clinical practice: Beginning from 1st April 2021 the application was used as an additional patient screening tool for the four trials CLOSURE-AF, FAIR-HF2, SPRIRIT-HF and TORCH-PLUS of the German Centre for Cardiovascular Research. As the COVID-19 pandemic is prohibiting any proper comparison of patient inclusion rates, efficacy of the recruitment support system was tested by comparing the numbers of patients identified by the recruitment support system and enrolled in a trial to the actual number of enrolled patients irrespective of the screening method from 1st April 2021 to 23rd November 2021. The system was able to identify 52 of 55 patients included in those four clinical trials. Conclusion Use of FHIR for defining eligibility criteria of clinical trials may facilitate interoperability and allow automatic screening for eligible patients at multiple sites of different healthcare providers in the future. Upcoming changes in FHIR should allow easier description of “OR”-linked eligibility criteria. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Deutsche Forschungsgemeinschaft

BUPRENORPHINE WAIVER TRAINING: PREPARING CHILD AND ADOLESCENT PSYCHIATRISTS TO TREAT OPIOID USE DISORDERS

This Workshop will prepare child and adolescent psychiatrists to integrate buprenorphine into their practice by providing the necessary training to apply for a Drug Enforcement Administration (DEA) waiver to prescribe buprenorphine to more than 30 patients. The Workshop will specifically emphasize implementation with discussions around solutions to commonly identified barriers to buprenorphine prescribing, and linkage to resources for future free mentorship through the Provider Clinical Support System (funded by SAMHSA).

Expérience sensorielle et affective du lien dans le cadre du psychotraumatisme. Médiation thérapeutique par le théâtre

ObjectivesThis article retraces part of the accompaniment of a ten-year-old boy in a special education program. This article raises questions about the child's behavior, which oscillated between fury and tranquillity, as if there was only one step between the two. The authors focus on the functional particularity of psychotrauma which, by its weight, splits certain psychic zones from the rest of the sensory world. They try to understand the link, in its sensory and affective dimension, between areas that no longer participate in the subject's psychic life and the rest of the sensory world, within the precise framework of psychotrauma. MethodThe authors combine art therapy techniques with a psychoanalytical approach, combining the art of theater with the metaphorical capacity offered by Dixit cards. This particular theater, called “Dixian Theatre,” offers a place of scenic expression where the appearance of fury is not stopped, but delimited. ResultsIn this space, and through the accompaniment of the therapists, it appears that the violence of this young boy manifests itself on stage; the child is not in the grip of furor, and the violence that he enacts is bearable for all. The child is gradually able to link his present moments of fury to the terrible events of his past. DiscussionThe device proposed by the authors gives rise to a particular transferential relationship where the clinician becomes the place of deposit of a psychic reality that could not have been elaborated until that moment. The therapist then becomes the link of an associative chain promoting the restitution of psychic content that the child can reappropriate. ConclusionThe clinical support system encourages bodily expression and the adult's reverie. The theater, which became therapeutic, is able to help relaunch a psychic movement, after certain contents had become inaccessible and bruised by trauma. The child can then regain access to a temporal dynamic mixing a “before,” a “during,” and an “after” trauma.

FINDINGS AND FUTURE DIRECTIONS FROM A SMOKING CESSATION TRIAL UTILIZING A CLINICAL DECISION SUPPORT TOOL.

Tobacco smoking is the leading cause of disease, death, and disability in the United States. Dental practitioners are advised to provide evidence-based smoking cessation interventions to their patients, yet dental practitioners frequently fail to deliver brief smoking cessation advice. To test whether giving dental practitioners a clinical decisions support (CDS) system embedded in their electronic dental record would increase the rate at which patients who smoke (1) report receiving a brief intervention or referral to treatment during a recent dental visit, (2) taking action related to smoking cessation within 7 days of visit, and (3) stop smoking for 1 day or more or reduce the amount smoked by 50% within 6 months. Two-group, parallel arm, cluster-randomized trial. From March through December 2019, 15 nonacademic primary care dental clinics were randomized via covariate adaptive randomization to either a usual care arm or the CDS arm. Adult smokers completed an initial telephone survey within 7 days of their visit and another survey after 6 months. Forty-three patients from 5 CDS and 13 patients from 2 usual care clinics completed the 7-day survey. While the proportion of patients who reported receipt of a brief intervention or referral to treatment was significantly greater in the CDS arm than the usual care arm (84.3% vs 58.6%; P = .005), the differences in percentage of patients who took any action related to smoking cessation within 7 days (44.4% vs 22.3%; P = .077), or stopped smoking for one day or more and/or reduced amount smoked by 50% within 6 months (63.1% vs 46.2%; P = .405) were large but not statistically significant. Despite interruption by COVID-19, these results demonstrate a promising approach to assist dental practitioners in providing their patients with smoking cessation screening, brief intervention and referral to treatment.

Screening for Heart Disease in the Age of Digital Health Technologies: Who, When, and How?

Heart disease affects much of the world’s population, yet many people have no idea that they could have something wrong with them. An opportunity therefore exists for targeted screening for conditions such as cardiovascular disease, heart rhythm changes, valvular heart disease, structural abnormalities, and more subtle, rarer inherited heart conditions. At the same time, the rapid development of digital health technologies and clinical support systems is providing patients and their doctors access to augmented intelligence solutions to diagnose these conditions. This article will focus on how the emerging field of digital health technology can aid screening for heart disease and explore its usefulness in disease specific and population specific groups.

Implementation Outcomes Assessment of a Digital Clinical Support Tool for Intrapartum Care in Rural Kenya: Observational Analysis.

BackgroundiDeliver, a digital clinical support system for maternal and neonatal care, was developed to support quality of care improvements in Kenya.ObjectiveTaking an implementation research approach, we evaluated the adoption and fidelity of iDeliver over time and assessed the feasibility of its use to provide routine Ministry of Health (MOH) reports.MethodsWe analyzed routinely collected data from iDeliver, which was implemented at the Transmara West Sub-County Hospital from December 2018 to September 2020. To evaluate its adoption, we assessed the proportion of actual facility deliveries that was recorded in iDeliver over time. We evaluated the fidelity of iDeliver use by studying the completeness of data entry by care providers during each stage of the labor and delivery workflow and whether the use reflected iDeliver’s envisioned function. We also examined the data completeness of the maternal and neonatal indicators prioritized by the Kenya MOH.ResultsA total of 1164 deliveries were registered in iDeliver, capturing 45.31% (1164/2569) of the facility’s deliveries over 22 months. This uptake of registration improved significantly over time by 6.7% (SE 2.1) on average in each quarter-year (P=.005), from 9.6% (15/157) in the fourth quarter of 2018 to 64% (235/367) in the third quarter of 2020. Across iDeliver’s workflow, the overall completion rate of all variables improved significantly by 2.9% (SE 0.4) on average in each quarter-year (P<.001), from 22.25% (257/1155) in the fourth quarter of 2018 to 49.21% (8905/18,095) in the third quarter of 2020. Data completion was highest for the discharge-labor summary stage (16,796/23,280, 72.15%) and lowest for the labor signs stage (848/5820, 14.57%). The completion rate of the key MOH indicators also improved significantly by 4.6% (SE 0.5) on average in each quarter-year (P<.001), from 27.1% (69/255) in the fourth quarter of 2018 to 83.75% (3346/3995) in the third quarter of 2020.ConclusionsiDeliver’s adoption and data completeness improved significantly over time. The assessment of iDeliver’ use fidelity suggested that some features were more easily used because providers had time to enter data; however, there was low use during active childbirth, which is when providers are necessarily engaged with the woman and newborn. These insights on the adoption and fidelity of iDeliver use prompted the team to adapt the application to reflect the users’ culture of use and further improve the implementation of iDeliver.

A Multi-View Learning based Clustering Method for Health Care System

Abstract: Patients in hospitals have faced complications due to measurement errors, missing data, privacy issues etc. in electronic medical records. However, these medical records from heterogeneous sources have both structured and unstructured data. In particular, unstructured clinical data is valuable source of information including patient’s records of pathology data, radiology findings, medication order etc. However, to scrutinize, construe and presentation of this unstructured and high dimensional data is one of the significant modeling challenge that clinical support system has faced from many years before. Therefore, there is a need of some standard technique to locate both subjective and objective guesstimates of patient’s condition. Our endowments in this paper are twofold. First, present a multi-view learning technique, i.e. Collective Matrix Factorization to combine the extracted features from multiple views and gives a low dimensional representation of combined clinical data. Second, proposed a Genetic-K-means based clustering algorithm based on Collective Matrix Factorization for heterogeneous clinical records. It has been observed by the experiments that proposed method gives more accurate clustering results than existing method. Keywords: Clinical notes; Collective Matrix Factorization; Genetic; heterogeneous data; K-means; Multi-view learning.

Implementation of a Clinical Trial Recruitment Support System Based on Fast Healthcare Interoperability Resources (FHIR) in a Cardiology Department.

Clinical Trial Recruitment Support Systems can booster patient inclusion of clinical trials by automatically analyzing eligibility criteria based on electronic health records. However, missing interoperability has hindered introduction of those systems on a broader scale. Therefore, our aim was to develop a recruitment support system based on FHIR R4 and evaluate its usage and features in a cardiology department. Clinical conditions, anamnesis, examinations, allergies, medication, laboratory data and echocardiography results were imported as FHIR resources. Clinical trial information, eligibility criteria and recruitment status were recorded using the appropriate FHIR resources without extensions. Eligibility criteria linked by the logical operation "OR" were represented by using multiple FHIR Group resources for enrollment. The system was able to identify 52 of 55 patients included in four clinical trials. In conclusion, use of FHIR for defining eligibility criteria of clinical trials may facilitate interoperability and allow automatic screening for eligible patients at multiple sites of different healthcare providers in the future. Upcoming changes in FHIR should allow easier description of "OR"-linked eligibility criteria.

Patient Data Sharing and Reduction of Overtime Work of Nurses by Innovation of Nursing Records Using Structured Clinical Knowledge.

Half of nurses' overtime hours are due to records. Nursing records, which are mainly narrative records, cost a large amount of money. However, it has been pointed out that there are problems with their quality and post-use. In this study, we analyzed the value of nursing records for physicians. As a result, we found that the use of standard observation terms in nursing records can create an environment in which patients' conditions can be shared. To create this environment, the physicians of the clinical path committee classified hospitalized patients in terms of disease, treatment, and examination, and created a list of 778 process paths. Physicians, nurses, and researchers collaborated to develop digital contents with high-priority observation items and care actions adapted to patient conditions for each path. We developed a clinical support system equipped with these digital contents. In May 2019, we installed the system in a 900-bed university hospital. Then, in October 2020, we installed the system in a 400-bed general hospital. We used "nurses' overtime hours for recording" and "reduction rate" as indicators of the usefulness of this system. In the 900-bed university hospital, we compared the previous year's results for March, the end of the fiscal year. This overtime hours were 2,944 hours 00 minutes in March 2019 and 2,141 hours 55 minutes in March 2020. 27% reduction was indicated. The respective bed occupancy rates were 90.80 percent and 90.60 percent, with no difference. In the 400-bed general hospital, This overtime hours were compared to the previous year, covering November and December after one month of implementation. 386 hours in November 2019 and 204.5 hours in November 2020. 47% reduction indicated. 366 hours in December 2019 and 214.5 hours in December 2020. A reduction of 41% was shown. These results suggest that the implementation of this system can both improve the quality of team care and reduce overtime.

Artificial intelligence algorithms for the recognition of Brugada type 1 pattern on standard 12-leads ECG

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): This research project is funded by Tuscany Region Background/Introduction Electrocardiograms (ECGs) are rapidly moving from analog to digital versions. Consequently, a series of automatic analyses of standard 12-lead ECGs are attracting interest for their ability to support clinicians in the automatic recognition of specific features associated with different cardiac diseases [2]. Artificial Intelligence applications and Machine Learning (ML) algorithms have gained much attention in the last years for their ability to figure out patterns from data independently, without being explicitly taught rules. Peculiar features define the ECGs of patients with Brugada Syndrome (BrS); however, ambiguities still exist for the correct diagnosis of BrS and discrimination with respect to other pathologies. Purpose The BrAID (Brugada syndrome and Artificial Intelligence applications to Diagnosis) project aims to develop an innovative system for diagnosing Type 1 BrS based on ECG pattern recognition through the application of ML algorithms. In this work, an application of Echo State Networks (ESN), a type of Recurrent Neural Network (RNN), for the diagnosis of BrS from ECG is presented. Methods After approval from the Local Ethical Committees, 12-lead ECGs were obtained in patients enrolled in 5 Centers diagnosed with typical spontaneous Type 1 pattern (coved) (group A, 81 patients). Baseline ECG was also collected in patients undergoing the ajmaline test, classified as positive (group B, 37 patients) or negative (group C, 14 patients) according to test results. 174 patients with no clinical and familial history of arrhythmias were considered controls (group D). Data were collected from 4 beats extracted from the ECGs as input to the ESN. The datasets obtained in the different groups were used for the ESN model’s training and assessment (testing) through a double cross-validation approach. Results As shown in Table 1, the performances using three leads (V1, V2, V3) or V2 only were compared. The algorithm performance was assessed in all the datasets (group A+B+C+D) and in spontaneous BrS (group A) and controls (group D). A good accuracy (79.21%) was seen when the three leads were considered for groups A and D only; the best test set accuracy (80.20%) was obtained in the case in which V2 only was used as input in all the datasets. Conclusion(s) In this work, a novel system for diagnosing Type 1 BrS using an ESN approach was developed. Our preliminary results show that this ML model is able to detect ECG patterns associated with Type 1 BrS with good and comparable accuracy both when three leads (79.21% ) or V2 only (80.20%) were analyzed. The future availability of larger datasets could improve the model performance, increasing the ESN potentialities as a clinical support system tool to be used in everyday clinical practice. Table 1. The accuracy, specificity, and sensitivity reported for each dataset group are obtained through double cross-validation.

Safety assessment of a redirection program using an electronic application for low-acuity patients visiting an emergency department

BackgroundEmergency departments (EDs) are operating at or above capacity, which has negative consequences on patients in terms of quality of care and morbi-mortality. Redirection strategies for low-acuity ED patients to primary care practices are usually based on subjective eligibility criteria that sometimes necessitate formal medical assessment. Literature investigating the effect of those interventions is equivocal. The aim of the present study was to assess the safety of a redirection process using an electronic clinical support system used by the triage nurse without physician assessment.MethodsA single cohort observational study was performed in the ED of a level 1 academic trauma center. All low-acuity patients redirected to nearby clinics through a clinical decision support system (February–August 2017) were included. This system uses different sets of medical prerequisites to identify patients eligible to redirection. Data on safety and patient experience were collected through phone questionnaires on day 2 and 10 after ED visit. The primary endpoint was the rate of redirected patients returning to any ED for an unexpected visit within 48 h. Secondary endpoints were the incidence of 7-day return visit and satisfaction rates.ResultsA total of 980 redirected low-acuity patients were included over the period: 18 patients (2.8%) returned unexpectedly to an ED within 48 h and 31 patients (4.8%) within 7 days. No hospital admission or death were reported within 7 days following the first ED visit. Among redirected patients, 81% were satisfied with care provided by the clinic staff.ConclusionThe implementation of a specific electronic-guided decision support redirection protocol appeared to provide safe deferral to nearby clinics for redirected low-acuity patients. EDs are pivotal elements of the healthcare system pathway and redirection process could represent an interesting tool to improve the care to low-acuity patients.

IoT-Based Health Monitoring System Development and Analysis

This paper presents the design and implementation of a health monitoring system using the Internet of Things (IoT). In present days, with the expansion of innovations, specialists are always looking for innovative electronic devices for easier identification of irregularities within the body. IoT-enabled technologies enable the possibility of developing novel and noninvasive clinical support systems. This paper presents a health care monitoring system. In particular, COVID-19 patients, high blood pressure patients, diabetic patients, etc., in a rural area in a developing country, such as Bangladesh, do not have instant access to health or emergency clinics for testing. Buying individual instruments or continuous visitation to hospitals is also expensive for the regular population. The system we developed will measure a patient’s body temperature, heartbeat, and oxygen saturation (SpO2) levels in the blood and send the data to a mobile application using Bluetooth. The mobile application was created via the Massachusetts Institute of Technology (MIT) inventor app and will receive the data from the device over Bluetooth. The physical, logical, and application layers are the three layers that make up the system. The logical layer processes the data collected by the sensors in the physical layer. Media access management and intersensor communications are handled by the logical layer. Depending on the logical layer’s processed data, the application layer makes decisions. The main objective is to increase affordability for regular people. Besides sustainability in the context of finance, patients will have easy access to personal healthcare. This paper presents an IoT-based system that will simplify the utilization of an otherwise complicated medical device at a minimum cost while sitting at home. A 95 percent confidence interval with a 5 percent maximum relative error is applied to all measurements related to determining the patient’s health parameters. The use of these devices as support tools by the general public in a certain situation could have a big impact on their own lives.

Responsible AI in Healthcare: Best Practices to Improve Healthcare Delivery

A candid discussion that dives into the recesses of AI and ethics including topics such as what lessons can be used in healthcare from tech players implementing responsible AI and unbiased systems, how can responsible AI help build a clinical support system, examples from providers and payers and learnings from responsible AI which can be used in Obamacare and clinical trial implementation.&nbsp;&nbsp;&nbsp; Discussion topics will include: What is responsible AI? Is it the same&nbsp;as Ethical AI?&nbsp; What lessons can be used in healthcare from tech players implementing responsible AI and unbiased systems? Why is responsible AI important for healthcare? Is there demographic bias in the treatment part of responsible AI? How can responsible AI help build a clinical support system?&nbsp; What are some examples of responsible AI in the healthcare industry with respect to providers and payers?&nbsp; What are learnings from responsible AI which can be used in Obamacare implementation? HHS introduced an&nbsp;interoperability&nbsp;rule between EMR/EHR systems in 2019. How does it help in the implementation of responsible AI? &nbsp; How responsible AI is currently covered in new US government initiative such as&nbsp;AI.gov Responsible AI &amp; data bias in clinical trials&nbsp;

Initial Responses to False Positives in AI-Supported Continuous Interactions: A Colonoscopy Case Study

The use of artificial intelligence (AI) in clinical support systems is increasing. In this article, we focus on AI support for continuous interaction scenarios. A thorough understanding of end-user behaviour during these continuous human-AI interactions, in which user input is sustained over time and during which AI suggestions can appear at any time, is still missing. We present a controlled lab study involving 21 endoscopists and an AI colonoscopy support system. Using a custom-developed application and an off-the-shelf videogame controller, we record participants’ navigation behaviour and clinical assessment across 14 endoscopic videos. Each video is manually annotated to mimic an AI recommendation, being either true positive or false positive in nature. We find that time between AI recommendation and clinical assessment is significantly longer for incorrect assessments. Further, the type of medical content displayed significantly affects decision time. Finally, we discover that the participant’s clinical role plays a large part in the perception of clinical AI support systems. Our study presents a realistic assessment of the effects of imperfect and continuous AI support in a clinical scenario.

Disambiguating Clinical Abbreviations Using a One-Fits-All Classifier Based on Deep Learning Techniques.

Background Abbreviations are considered an essential part of the clinical narrative; they are used not only to save time and space but also to hide serious or incurable illnesses. Misreckoning interpretation of the clinical abbreviations could affect different aspects concerning patients themselves or other services like clinical support systems. There is no consensus in the scientific community to create new abbreviations, making it difficult to understand them. Disambiguate clinical abbreviations aim to predict the exact meaning of the abbreviation based on context, a crucial step in understanding clinical notes. Objectives Disambiguating clinical abbreviations is an essential task in information extraction from medical texts. Deep contextualized representations models showed promising results in most word sense disambiguation tasks. In this work, we propose a one-fits-all classifier to disambiguate clinical abbreviations with deep contextualized representation from pretrained language models like Bidirectional Encoder Representation from Transformers (BERT). Methods A set of experiments with different pretrained clinical BERT models were performed to investigate fine-tuning methods on the disambiguation of clinical abbreviations. One-fits-all classifiers were used to improve disambiguating rare clinical abbreviations. Results One-fits-all classifiers with deep contextualized representations from Bioclinical, BlueBERT, and MS_BERT pretrained models improved the accuracy using the University of Minnesota data set. The model achieved 98.99, 98.75, and 99.13%, respectively. All the models outperform the state-of-the-art in the previous work of around 98.39%, with the best accuracy using the MS_BERT model. Conclusion Deep contextualized representations via fine-tuning of pretrained language modeling proved its sufficiency on disambiguating clinical abbreviations; it could be robust for rare and unseen abbreviations and has the advantage of avoiding building a separate classifier for each abbreviation. Transfer learning can improve the development of practical abbreviation disambiguation systems.

Advantages and Functions of Clinical and Decision Support Systems

Clinical Decision Support Systems (CDSSs) signify the framework shift in the medical sector in the modern age. CDSSs are utilized in augmenting healthcare facilities in the process of making complex clinical decisions. Since the first application of CDSSs in the 80s, the framework has witnessed significant transformation. The frameworks are now administered through electronic healthcare records with complex capacities. Irrespective of these complex advancements, there are existing questions concerning the impacts of CDSSs on service providers, healthcare costs, and patients’ records. There are many published texts concerning the success stories of CDSSs, but significant setbacks have proved that CDSSs are not without any potential risks. In this research, we provide critical analysis on the application of CDSSs in clinical setting, integrating various forms, present application cases with proven effectiveness, potential harms and common pitfalls. We therefore conclude with evidence-centered recommendation for mitigating the issues of CDSSs maintainability, evaluation, implementation and designing.

ColonFormer: An Efficient Transformer Based Method for Colon Polyp Segmentation

Identifying polyps is challenging for automatic analysis of endoscopic images in computer-aided clinical support systems. Models based on convolutional networks (CNN), transformers, and their combinations have been proposed to segment polyps with promising results. However, those approaches have limitations either in modeling the local appearance of the polyps only or lack of multi-level feature representation for spatial dependency in the decoding process. This paper proposes a novel network, namely ColonFormer, to address these limitations. ColonFormer is an encoder-decoder architecture capable of modeling long-range semantic information at both encoder and decoder branches. The encoder is a lightweight architecture based on transformers for modeling global semantic relations at multi scales. The decoder is a hierarchical network structure designed for learning multi-level features to enrich feature representation. Besides, a refinement module is added with a new skip connection technique to refine the boundary of polyp objects in the global map for accurate segmentation. Extensive experiments have been conducted on five popular benchmark datasets for polyp segmentation, including Kvasir, CVC-Clinic DB, CVC-ColonDB, CVC-T, and ETIS-Larib. Experimental results show that our ColonFormer outperforms other state-of-the-art methods on all benchmark datasets.

APA Joins in Negotiations Over New Problematic Clozapine REMS

APA Joins in Negotiations Over New Problematic Clozapine REMS