Qualitative Evaluation of an Artificial Intelligence-Based Clinical Decision Support System to Guide Rhythm Management of Atrial Fibrillation: Survey Study.

BACKGROUND: Despite the numerous studies evaluating various rhythm control strategies for atrial fibrillation (AF), determination of the optimal strategy in a single patient is often based on trial and error, with no one-size-fits-all approach based on international guidelines/recommendations. The decision, therefore, remains personal and lends itself well to help from a clinical decision support system, specifically one guided by artificial intelligence (AI). QRhythm utilizes a 2-stage machine learning (ML) model to identify the optimal rhythm management strategy in a given patient based on a set of clinical factors, in which the model first uses supervised learning to predict the actions of an expert clinician and identifies the best strategy through reinforcement learning to obtain the best clinical outcome-a composite of symptomatic recurrence, hospitalization, and stroke. OBJECTIVE: We qualitatively evaluated a novel, AI-based, clinical decision support system (CDSS) for AF rhythm management, called QRhythm, which uses both supervised and reinforcement learning to recommend either a rate control or one of 3 types of rhythm control strategies-external cardioversion, antiarrhythmic medication, or ablation-based on individual patient characteristics. METHODS: Thirty-three clinicians, including cardiology attendings and fellows and internal medicine attendings and residents, performed an assessment of QRhythm, followed by a survey to assess relative comfort with automated CDSS in rhythm management and to examine areas for future development. RESULTS: The 33 providers were surveyed with training levels ranging from resident to fellow to attending. Of the characteristics of the app surveyed, safety was most important to providers, with an average importance rating of 4.7 out of 5 (SD 0.72). This priority was followed by clinical integrity (a desire for the advice provided to make clinical sense; importance rating 4.5, SD 0.9), backward interpretability (transparency in the population used to create the algorithm; importance rating 4.3, SD 0.65), transparency of the algorithm (reasoning underlying the decisions made; importance rating 4.3, SD 0.88), and provider autonomy (the ability to challenge the decisions made by the model; importance rating 3.85, SD 0.83). Providers who used the app ranked the integrity of recommendations as their highest concern with ongoing clinical use of the model, followed by efficacy of the application and patient data security. Trust in the app varied; 1 (17%) provider responded that they somewhat disagreed with the statement, "I trust the recommendations provided by the QRhythm app," 2 (33%) providers responded with neutrality to the statement, and 3 (50%) somewhat agreed with the statement. CONCLUSIONS: Safety of ML applications was the highest priority of the providers surveyed, and trust of such models remains varied. Widespread clinical acceptance of ML in health care is dependent on how much providers trust the algorithms. Building this trust involves ensuring transparency and interpretability of the model.

The efficacy of low dose short-term prednisone therapy for remission induction in newly diagnosed rheumatoid arthritis patients.

BACKGROUND: The ACR/EULAR recommendations endorse the use of glucocorticoids (GCs) for rheumatoid arthritis (RA) patients' flares and as a bridge to a DMARD. However, the recommendation of low dose short-term monotherapy with (GCs) remains open to the discretion of the clinician. The aim of this study was to assess whether a short-term use of low dose prednisone monotherapy was effective in inducing remission in newly diagnosed RA patients. METHODS: A retrospective analysis of patients newly diagnosed with RA at a Community Health Center in North Dakota was performed based on the ACR/EULAR RA classification criteria. Demographic and clinical data were abstracted from patients' medical charts. Patients treated with (< 10 mg/day) of prednisone up to 6 months were included. Response to prednisone was analyzed according to pre- and post-treatment DAS28-ESR score and EULAR response criteria. RESULTS: Data on 201 patients were analyzed. The mean prednisone dose was 8 mg/day (range: 5-10; SD = 1.2) and the mean treatment duration was 42.2 days (12-177; 16.9). Disease severity significantly improved from baseline to follow-up for: tender joint count (8.6 ± 4.8 vs. 1.5 ± 3.3; P < 0.001), swollen joint count (6.2 ± 5.0 vs. 1.4 ± 3.0; P < 0.001), and visual analog pain score (4.8 ± 2.6 vs. 2.1 ± 2.5; P < 0.001). DAS28-ESR disease severity significantly improved from baseline to follow-up: (5.1 ± 1.2 vs. 2.7 ± 1.3; P < 0.001). Per EULAR response criteria, 69.7% of patients showed good response to treatment and 20.4% showed moderate response. 54.2% of patients reached remission. CONCLUSION: Short-term use of low dose prednisone monotherapy induced disease remission and improved clinical severity of RA in the majority of newly diagnosed patients.

The efficacy of low dose short-term prednisone therapy for remission induction in newly diagnosed rheumatoid arthritis patients

Abstract Background: The ACR/EULAR recommendations endorse the use of glucocorticoids (GCs) for rheumatoid arthritis (RA) patients' flares and as a bridge to a DMARD. However, the recommendation of low dose short-term monotherapy with (GCs) remains open to the discretion of the clinician. The aim of this study was to assess whether a short-term use of low dose prednisone monotherapy was effective in inducing remission in newly diagnosed RA patients. Methods: A retrospective analysis of patients newly diagnosed with RA at a Community Health Center in North Dakota was performed based on the ACR/EULAR RA classification criteria. Demographic and clinical data were abstracted from patients' medical charts. Patients treated with (≤ 10 mg/day) of prednisone up to 6 months were included. Response to prednisone was analyzed according to pre- and post-treatment DAS28-ESR score and EULAR response criteria. Results: Data on 201 patients were analyzed. The mean prednisone dose was 8 mg/day (range: 5-10; SD = 1.2) and the mean treatment duration was 42.2 days (12-177; 16.9). Disease severity significantly improved from baseline to follow-up for: tender joint count (8.6 ± 4.8 vs. 1.5 ± 3.3; P < 0.001), swollen joint count (6.2 ± 5.0 vs. 1.4 ± 3.0; P < 0.001), and visual analog pain score (4.8 ± 2.6 vs. 2.1 ± 2.5; P < 0.001). DAS28-ESR disease severity significantly improved from baseline to follow-up: (5.1 ± 1.2 vs. 2.7 ± 1.3; P < 0.001). Per EULAR response criteria, 69.7% of patients showed good response to treatment and 20.4% showed moderate response. 54.2% of patients reached remission. Conclusion: Short-term use of low dose prednisone monotherapy induced disease remission and improved clinical severity of RA in the majority of newly diagnosed patients.

DAS28-CRP Cutoffs for High Disease Activity and Remission Are Lower Than DAS28-ESR in Rheumatoid Arthritis.

OBJECTIVE: Guidelines do not specify how cutoffs for high disease activity differ between the Disease Activity Score 28-joint count indices DAS28-erythrocyte sedimentation rate (ESR) and DAS28-C-reactive protein (CRP). Studies that compare DAS28-CRP and DAS28-ESR depend on data from clinical trials, registries, or practices with multiple providers. Existing studies use data from patients who received immunosuppressive therapy. This study compared the DAS28-ESR and DAS28-CRP values from immunosuppressive treatment-naïve patients in a single physician practice. METHODS: A retrospective electronic medical chart review was conducted for new diagnoses of rheumatoid arthritis (RA; International Classification of Diseases [ICD]-9 714), based on the American College of Rheumatology/European League against Rheumatology 2010 RA classification criteria. The number of patients with high disease activity (>5.1) was compared using ESR and CRP data to calculate the proportion of discordance. A receiver operator curve and Youden index was used to calculate the DAS28-CRP high disease activity cutoff estimation that corresponds with DAS28-ESR of more than 5.1. RESULTS: There were 171 patients included in this study. The baseline mean DAS28-ESR was higher than the baseline mean DAS-28 CRP: 5.1 ± 1.2 versus 4.1 ± 1.0 (P < 0.001); 48.5% of patients met criteria for high disease activity (score >5.1) compared with only 14.6% when measured by DAS28-CRP. Discordance was 33.9%. κ coefficient was only .307. Receiver operator curve and Youden index analysis suggested that the cutoff point for high disease activity of DAS28-CRP greater than 4.1, which corresponds to DAS28-ESR greater than 5.1. Similarly, DAS28-ESR posttreatment scores were significantly higher than DAS28-CRP. When measured by DAS28-ESR, patients in remission had higher scores as measured by DAS28-ESR (1.81) than DAS28-CRP (1.45). CONCLUSION: There is a difference between DAS28-ESR and DAS28-CRP, even when calculated for immunosuppressive treatment-naïve patients. DAS28-CRP is significantly lower than DAS28-ESR.

Pregnancy Status is Associated With Screening for Alcohol and Other Substance Use in the Emergency Department.

OBJECTIVES: Determine the rates of screening for substance use in pregnant women versus non-pregnant women attending the emergency department (ED). METHODS: We captured all ED visits by women of childbearing age (12-50 years in our study) over a 5-year period (2012-2017) (n = 72,752) from a local community hospital. The 2742 pregnant women were then matched by ethnicity, marital status, and arrival method to 9888 non-pregnant women. We then compared rates of screening for substance use by pregnancy status stratifying by age and diagnosis. RESULTS: The proportion of non-pregnant women who were screened for substance use was 3.66% compared to 1.90% of pregnant women, yielding an odds ratio (OR) of 1.96 (95% CI = 1.44 to 2.67). We then stratified the results by presenting complaint and age. Non-pregnant women 14 to 19 and 30 to 34 had the highest likelihood for screening (OR > 3.0). The presenting complaint showed little effect on screening. CONCLUSION: Pregnant women were screened only 51% as often as non-pregnant women for substance use in the ED. These results are of particular concern as we continue to see a rise in substance use during pregnancy which results in an increased burden on the healthcare system and society. This study replicates a previous study showing that the rates of screening are lower for pregnant women than non-pregnant women presenting to the ED. Earlier recognition of substance use offers increased opportunities for intervention and prevention of adverse outcomes from substance use during both the current pregnancy and future pregnancies.

Identification of prostate cancer antigens by automated high-throughput filter immunoscreening.

There is a need for earlier and more accurate cancer diagnostics as well as new targets for cancer immunotherapy. To this end, it is important to identify sets of tumour antigens specific for different cancer forms. Several methods that identify potential tumour antigens in an arrayed and high-throughput format have been developed during the last years of SEREX (serological identification of antigens by recombinant expression cloning) related research. Such techniques may hold the potential to describe the complete immunogenic part of the cancer proteome, also called the cancer immunoproteome. We have developed a powerful platform for automated serological high-throughput filter screening of tumour cDNA libraries. The screening format of this method is 18,000 single cDNAs clones, which is superior to other high-throughput methods described. The output is antigens, which are potential diagnostic cancer markers and vaccine targets. We present here the results from the screening of a prostate tumour cDNA library with autologous patient antibodies. We first demonstrated the feasibility of the automated high-throughput filter immunoscreening method by use of the NY-ESO-1sv (NY-ESO-1 splice variant) antigen. We then screened 18,000 cDNA clones from a phage display selected prostate tumour cDNA library with autologous patient antibodies and identified several relevant antigens such as NY-ESO-1, XAGE-1, DJ-1 and transcription factor 25 (TCF25). The present high-throughput immunoscreening method has the potential to identify both patient-specific and disease-specific antigens for use in diagnostics and therapy.

Construction, evaluation and refinement of a large human antibody phage library based on the IgD and IgM variable gene repertoire.

The ability to isolate antibodies against any antigen of interest has become increasingly important as antibodies have proved their utility both in antigen detection, quantification and as specific in vivo targeting agents. To this end, we have constructed a large antibody phage library in the single chain Fv (scFv) phagemid format based on the naive human variable (V) gene repertoire dictated by IgD and IgM. Optimizing each step of the library construction has resulted in a highly diverse and functional library, as assessed by sequencing analysis, large-scale automated expression analysis and antigen screening. Furthermore, the versatile format of the library, which comprises 14 separate sub-libraries, adds considerably flexibility with respect to which part of the antibody repertoire that is to be probed. This versatility has been further exploited to generate a refined antibody library, which exhibits one of the highest prokaryotic expression levels reported to date for a naive repertoire. The construction of the refined library was based on the functional purification of expressed V genes in the context of the protein L interaction with correctly folded V genes of the kappa light chain family. Antigen screening of this library indicated that the functional purification improved the ability to retrieve antigen specific antibodies, but at the cost of potential loss of diversity in the isolated repertoire.

Covalent antibody display--an in vitro antibody-DNA library selection system.

The endonuclease P2A initiates the DNA replication of the bacteriophage P2 by making a covalent bond with its own phosphate backbone. This enzyme has now been exploited as a new in vitro display tool for antibody fragments. We have constructed genetic fusions of P2A with single-chain antibodies (scFvs). Linear DNA of these fusion proteins were processed in an in vitro coupled transcription-translation mixture of Escherichia coli S30 lysate. Complexes of scFv-P2A fusion proteins covalently bound to their own DNA were isolated after panning on immobilized antigen, and the enriched DNAs were recovered by PCR and prepared for the subsequent cycles of panning. We have demonstrated the enrichment of scFvs from spiked libraries and the specific selection of different anti-tetanus toxoid scFvs from a V-gene library with 50 million different members prepared from human lymphocytes. This covalent antibody display technology offers a complete in vitro selection system based exclusively on DNA-protein complexes.

Direct isolation of recombinant human antibodies against group B Neisseria meningitidis from scFv expression libraries.

The successful generation of human antibodies from large nai;ve antibody libraries requires iterative selection steps. Here, we describe a new and fast method for the isolation of high affinity antibodies directly from human single chain Fv antibody (scFv) expression libraries. Escherichia coli scFv expression libraries were made from peripheral blood lymphocytes from four individuals vaccinated with group B Neisseria meningitidis outer membrane vesicle (OMV) vaccine. Forty thousand clones were directly screened for antibodies binding N. meningitidis strain 44/76 (B:15:P1.7,16). Of 430 specific clones detected, 225 candidates were isolated and re-screened against the N. meningitidis strains NZ-98/254 (B:4:P1.7b,4) giving 4% cross-reactive clones. Antibodies were further characterized by DNA sequencing, ELISA and surface plasmon resonance (SPR) analysis, showing broad V-gene diversity and nanomolar scFv affinities. Antibodies derived by this method may assist in the discovery and development of new vaccine antigens as well as therapeutic antibody agents for the treatment of meningococcal diseases.

Identification of scFv antibody fragments that specifically recognise the heroin metabolite 6-monoacetylmorphine but not morphine.

Use of phage display of recombinant antibodies and large repertoire naïve antibody libraries for identifying antibodies of high specificity has been extensively reported. Nevertheless, there have been few reported antibodies to haptens that have originated from naïve antibody libraries with potential use in diagnostics. We have used chain shuffling of lead single-chain fragment variable (scFv) antibodies, isolated from a naïve antibody library, to screen for antibodies that specifically recognise the major metabolite of heroin, 6-monoacetylmorphine (6MAM). The antibodies were identified by screening high-density colonies of Escherichia coli expressing soluble scFv antibody fragments without prior expression on bacteriophage (phage display). The antibodies recognise 6MAM with affinities of 1-3x10(-7) M with no crossreactivity to morphine. These antibodies can potentially be used for developing a rapid immunoassay in drug-testing programs. To our knowledge, this is the first report of an antibody that distinguishes 6MAM from its de-acetylated form, morphine.