Artificial intelligence on COVID-19 pneumonia detection using chest xray images

Recent studies show the potential of artificial intelligence (AI) as a screening tool to detect COVID-19 pneumonia based on chest x-ray (CXR) images. However, issues on the datasets and study designs from medical and technical perspectives, as well as questions on the vulnerability and robustness of AI algorithms have emerged. In this study, we address these issues with a more realistic development of AI-driven COVID-19 pneumonia detection models by generating our own data through a retrospective clinical study to augment the dataset aggregated from external sources. We optimized five deep learning architectures, implemented development strategies by manipulating data distribution to quantitatively compare study designs, and introduced several detection scenarios to evaluate the robustness and diagnostic performance of the models. At the current level of data availability, the performance of the detection model depends on the hyperparameter tuning and has less dependency on the quantity of data. InceptionV3 attained the highest performance in distinguishing pneumonia from normal CXR in two-class detection scenario with sensitivity (Sn), specificity (Sp), and positive predictive value (PPV) of 96%. The models attained higher general performance of 91-96% Sn, 94-98% Sp, and 90-96% PPV in three-class compared to four-class detection scenario. InceptionV3 has the highest general performance with accuracy, F1-score, and g-mean of 96% in the three-class detection scenario. For COVID-19 pneumonia detection, InceptionV3 attained the highest performance with 86% Sn, 99% Sp, and 91% PPV with an AUC of 0.99 in distinguishing pneumonia from normal CXR. Its capability of differentiating COVID-19 pneumonia from normal and non-COVID-19 pneumonia attained 0.98 AUC and a micro-average of 0.99 for other classes.

Lipoprotein(a) Levels at Birth and in Early Childhood - The COMPARE Study

Background and objective High lipoprotein(a) is a genetically determined causal risk factor for cardiovascular disease and 20% of the adult population has high levels (i.e. >42 mg/dL, >88 nmol/L). We investigated whether early life lipoprotein(a) levels measured in cord blood may serve as a proxy for neonatal venous blood levels, whether lipoprotein(a) birth levels (i.e. cord or venous) predict levels later in life, and whether early life and parental levels correlate. Methods The COMPARE study is a prospective cohort study of newborns (N=450) from Copenhagen, Denmark including blood sampling of parents. Plasma lipoprotein(a) was measured in cord blood (N=402), neonatal venous blood (N=356), and at 2 (N=320) and 15 months follow-up (N=148) of infants, and in parents (N=705). Results Mean lipoprotein(a) levels were 2.2(95%CI:1.9-2.5), 2.4(2.0-2.7), 4.1(3.4-4.9), and 14.6(11.4-17.9) mg/dL in cord, neonatal venous, and 2- and 15-months venous samples, respectively. Lipoprotein(a) levels in cord blood correlated strongly with neonatal venous blood levels (R2=0.95, p<0.001) and neonatal levels correlated moderately with 2- and 15-months levels (R2=0.68 and 0.67, both p<0.001). Birth levels ≥90th percentile predicted lipoprotein(a) >42 mg/dL at 15 months with positive predictive values of 89% and 85% for neonatal venous and cord blood. Neonatal and infant levels correlated weakly with parental levels, most pronounced at 15 months (R2=0.22, p<0.001). Conclusions Lipoprotein(a) levels are low in early life, cord blood may serve as a proxy for neonatal venous blood, and birth levels ≥90th percentile can identify newborns at risk of developing high levels.

Interpretation of Clinical and Laboratory Findings of Patients with Lupus Nephritis According to the Results of Biopsy

Background: There are controversies regarding the diagnosis of lupus nephritis. Also, its clinical manifestations and severity are different from one patient to another. Objectives: The current study aimed to interpret clinical and laboratory features of lupus nephritis according to the results of the biopsy. Methods: Following a retrospective design, 30 patients with lupus, who were candidates for renal biopsy and undergoing kidney biopsy, were studied. Clinical findings (blood pressure and limb edema) and laboratory findings (Cr, ESR, CRP, BUN, C3, C4, CH50, Anti-ds DNA, and hematuria) were gathered. Finally, the diagnostic value of clinical and laboratory findings was interpreted according to the biopsies and the staging of samples in the pathology. Data were analyzed using SPSS. Quantitative variables are displayed using mean and quartiles. Fisher’s exact test was used to compare study groups. Also, independent samples t-test and Levene’s test were used to evaluate variances of quantitative variables. Results: Of 30 cases, 5 had a biopsy of 3 and FSGN. Also, 8 cases were on stages 3 and 4, 9 had a stage of 4 biopsy, and 8 cases were on stage 4 - 5, 5, and 5 - 6. The association between age, Limp edema, ESR, and biopsy was statistically significant (P < 0.05). Conclusions: This study demonstrated the usefulness of clinical and laboratory findings to determine the severity of the disease in the shortest time, mainly due to its easy, non-invasive access and early preparation of the results, which will facilitate the initiation of treatment.

POS1054 UPADACITINIB PHARMACOKINETICS AND EXPOSURE-RESPONSE RELATIONSHIPS FOR EFFICACY AND SAFETY IN PSORIATIC ARTHRITIS – ANALYSES OF THE PHASE 3 SELECT-PsA STUDIES

Background:Upadacitinib (UPA) is an oral, reversible, JAK inhibitor approved for the treatment of rheumatoid arthritis (RA). The efficacy and safety profile of UPA in psoriatic arthritis (PsA) has been established in the SELECT-PsA program which includes two global Phase 3 studies.Objectives:These analyses characterize UPA pharmacokinetics and exposure-response relationships for efficacy and safety endpoints using data from the SELECT-PsA studies.Methods:The SELECT-PsA program enrolled patients with prior inadequate response (IR) or intolerance to ≥1 non-bDMARD1 (N=1705) and prior IR or intolerance to ≥1 bDMARD2 (N=642). Data from both trials was integrated for patients receiving placebo (PBO), UPA 15mg once daily (QD) and UPA 30mg QD; adalimumab data was excluded from this analysis. UPA pharmacokinetics were characterized in PsA patients using Bayesian population pharmacokinetics analyses and utilizing prior information from analyses in healthy subjects and RA patients. Exposure-response analyses were conducted using logistic regression to characterize the relationships between upadacitinib average plasma concentration during a dosing interval (Cavg) and the percentage of patients achieving ACR20/50/70 at Weeks 12 and 24, static Investigator Global Assessment of psoriasis (sIGA) of 0 or 1 (clear or almost clear) and at least a 2-point improvement from baseline, and PASI75 at Weeks 16 and 24 or experiencing selected clinically relevant safety events through week 24.Results:Analyses were conducted using data from 1694 subjects (for pharmacokinetics) and 1916 subjects (for exposure-response analyses). UPA model-estimated plasma exposures in subjects with PsA who received 15mg and 30mg QD doses were comparable to previously estimated exposures in subjects with RA. Body weight and methotrexate use had no clinically relevant effects on UPA exposures. There was a statistically significant relationship between UPA Cavg and the percentage of subjects who achieved Week 12 ACR50/70, Week 16 sIGA 0/1, and Week 24 sIGA 0/1 (Figure 1). No statistically significant exposure-response relationship was observed for Week 12 ACR20, Week 16 PASI75, or Week 24 ACR20/50/70 or PASI75, indicating that the 15mg QD exposures are approximately at the plateau of response for these endpoints. No statistically significant relationships were observed between upadacitinib Cavg and the percentage of subjects experiencing pneumonia, herpes zoster, hemoglobin < 8 g/dL, Grade ≥3 lymphopenia, Grade ≥3 neutropenia. There was a shallow but statistically significant exposure-response relationships with the occurrence of serious infections and decrease in hemoglobin from baseline (>2 g/dL and >2 g/dL in combination with hemoglobin < lower limit for normal).Figure 1.Observed and Model Predicted Efficacy Responses at Week 12 (for ACR50/70) or at Weeks 16 and Week 24 (for sIGA 0/1) Versus Upadacitinib Plasma ExposuresConclusion:Exposure-response analyses demonstrated that plasma exposures associated with UPA 15 mg QD achieves robust efficacy in subjects with PsA with limited effects on the evaluated safety endpoints. UPA plasma exposures associated with UPA 15 and 30mg QD are predicted to provide similar ACR responses by week 24 while a small additional efficacy benefit with UPA 30mg is predicted for the achievement sIGA 0/1.References:[1]van Vollenhoven R, et al. Monotherapy with Upadacitinib in MTX-naïve Patients with Rheumatoid Arthritis: Results at 48 Weeks from the SELECT-EARLY Study. 2019 EULAR; THU0197[2]Fleischmann R, et al. Safety and Effectiveness of Upadacitinib or Adalimumab in Patients with Rheumatoid Arthritis: Results at 48 weeks from the SELECT-COMPARE Study. 2019 EULAR; FRI0147Disclosure of Interests:Elena Muensterman Shareholder of: AbbVie, Employee of: AbbVie, Benjamin Engelhardt Shareholder of: AbbVie, Employee of: AbbVie, Sathej Gopalakrishnan Shareholder of: AbbVie, Employee of: AbbVie, Jaclyn Anderson Shareholder of: AbbVie, Employee of: AbbVie, Mohamed-Eslam Mohamed Shareholder of: AbbVie, Employee of: AbbVie.

Comparison of Flowmetric Plethysmography and Forced Oscillatory Mechanics to Measure Airway Hyperresponsiveness in Horses

Airway hyperresponsiveness (AHR) is linked to airway inflammation and is considered a key manifestation of mild/moderate equine asthma (EA). The study purpose was to determine whether two modalities of non-invasive lung function testing (FOM—forced oscillatory mechanics vs. FP—flowmetric plethysmography) establish the same clinical diagnosis of AHR in horses, using histamine bronchoprovocation. Nineteen horses (3–25 years, 335–650 kg) with clinical signs suggestive of mild/moderate equine asthma were enrolled. FOM and FP testing was performed in each horse on two consecutive days, using a randomized cross-over design. AHR was defined by the histamine dose needed to double FOM baseline resistance, or to achieve a 35% increase in FP delta flow. Bronchoalveolar lavage fluid (BALF) was subsequently collected and stained with modified Wright's and toluidine blue stains. Binary statistical tests (related samples T-test, Mann-Whitney U, Chi-square analyses) were performed to compare study groups, with P &lt; 0.05 considered significant. Abnormal BALF cytology confirmed EA in 14/19 (73.7%) horses. Both FOM and FP revealed AHR in 7/14 (50%) of these EA horses. An additional 4/19 (21.1%) horses showed AHR based on FP but not FOM, including two horses with normal BALF cytology. A diagnosis of AHR was more often associated with FP than FOM (P = 0.013), although the prevalence of AHR was significantly higher in EA vs. non-EA horses, regardless of testing methodology. The phase angle between thoracic and abdominal components of breathing did not differ between test groups. In conclusion, FP diagnosed AHR more frequently than did FOM, including horses with no other diagnostic evidence of EA. Without further evaluation, these two testing modalities of AHR cannot be used interchangeably.

Measuring the performance of educational institute by using standard of total quality (compare study between private and public schools)

This research aims to Identify the Total Quality Management standards and the ability to implementation theses standards in educational institutes of Kurdistan region (KRG) in Public and Private schools by ask a question : what is the reality of implementation the standards of Total Quality Management (TQM) in educational institutes of (KRG). The study take 100 sample (Sava private school and AL-Khouwa public school) and use the excel program for analysis the data and the result showed there are difference in implementation the standards between the Public schools and Private schools. The study recommended the need of sharing and disseminating concepts of implementation the standards of (TQM) between the workers in the educational sector.

Clinical study on Sandhigata Vata w.r.s. to Osteoarthritis and its management by comparing the efficacy of Samananga Sneha and Samananga Sneha with Shadanga Guggulu

Sandhigata Vata is a disease seen commonly both in developing and developed countries. The disease causes mild, moderate or severe degree of morbidity and rarely mortality. The occurrence of problems is increasingly prevalent now a day due to change in lifestyle, food habits and social and cultural changes as well as travelling. Sandhishula, Sandhishotha, Sandhigraha and Atopa are the important clinical features of Sandhigata Vata. According to WHO, Osteoarthritis is the 2nd commonest problem in the world population i.e. 30%. The major risk factors associated with Kneejoint Osteoarthritis are Age, female sex, obesity, occupational Knee bending. Osteoarthritis is the most common articular disorder that begins asymptomatically and is extremely common by age 70. Almost all people by the age 40 have some pathologic change in weight bearing joints. 25% females and 16% males have symptomatic OA. To study the efficacy of Shamananga Sneha in Sandhigata Vata w.s.r. to Osteoarthritis knee. To compare (study) the efficacy of Shamananga Sneha with Shadanga Guggulu in Janu Sandhigata Vata. To compare the results of the two clinical groups. The study was comparative trial and 30 patients with Janu Sandhigata Vata. All the patients were randomly assigned into two groups namely A and B. Group A receive Shamananga Sneha (Guggulu Tiktaka Ghrita) for 30 days, and Group B receive Shamananga Sneha (Guggulu Tiktaka Ghrita) along with Shadanga Guggulu for 30 days. The patients were clinically evaluated and observations were recorded as in the case performa. Shamananga Sneha (Guggulu Tiktaka Ghrita) along with Shadanga Guggulu i.e. group B was more beneficial in relieving the symptoms than Shamananga Sneha i.e. group A (Guggulu Tiktaka Ghrita).

THU0201 LONG-TERM SAFETY AND EFFECTIVENESS OF UPADACITINIB OR ADALIMUMAB IN PATIENTS WITH RHEUMATOID ARTHRITIS: RESULTS AT 72 WEEKS FROM THE SELECT-COMPARE STUDY

Background:In the SELECT-COMPARE study in rheumatoid arthritis (RA) patients with inadequate response to methotrexate (MTX), upadacitinib (UPA), a Janus Kinase (JAK) 1-selective inhibitor, showed significant improvements in treatment of signs and symptoms when compared to placebo (PBO) and adalimumab (ADA) up to 48 weeks.1Objectives:To report safety and efficacy of UPA vs ADA up to 72 weeks in patients with RA from the ongoing long-term extension (LTE) of SELECT-COMPARE.Methods:Patients were randomized to once daily (QD) UPA 15 mg, PBO, or ADA 40 mg every other week, with all patients continuing background MTX. The study was double-blind for 48 weeks. Between Weeks 14-26, patients were rescued (from PBO to UPA, UPA to ADA, or ADA to UPA) if there was <20% improvement in tender/swollen joint count at Weeks 14/18/22 or if Clinical Disease Activity Index (CDAI) was >10 at Week 26; all PBO patients who were not rescued were switched to UPA at Week 26. Patients continued UPA or ADA in a blinded manner until the last patient completed the Week 48 visit; patients received open-label treatment thereafter. Study visits occurred at Week 60, 72, and every 12 weeks thereafter. Treatment-emergent adverse events (TEAEs) per 100 patient years (PY) were summarized up to December 26, 2018. Efficacy was analyzed by randomized group.Results:In total, 651, 651 and 327 patients were randomized at baseline to receive UPA, PBO, and ADA, respectively. Subsequently, 252 patients were switched from UPA to ADA, 159 were switched from ADA to UPA, and all PBO patients were switched to UPA. 1403 patients entered the LTE at Week 48 (UPA: 1091 [565 switched from PBO; 66 rescued from ADA; 460 on continued UPA]; ADA: 312 [110 rescued from UPA; 202 on continued ADA]). The cumulative exposures were 1396.7 and 515.1 PYs for UPA and ADA, respectively. UPA + MTX was generally well-tolerated as assessed by the frequency of AEs, including serious AEs, AEs leading to discontinuation of study drug, and AEs of special interest ([AESIs] including serious and opportunistic infections, malignancy, adjudicated major adverse cardiac events or venous thromboembolism; Figure 1). The event rates of AESIs were generally comparable between UPA + MTX and ADA + MTX, except for herpes zoster, lymphopenia, hepatic disorder, and CPK elevation, which were numerically higher with UPA + MTX. At both Weeks 60 and 72, significantly greater proportions of patient receiving UPA + MTX achieved ACR20/50/70 (P ≤.01/.001/.001), low disease activity (P ≤.001) and remission (P ≤.001) compared to those receiving ADA + MTX; Figure 2). Similarly, improvements in pain and function were significantly greater in the UPA vs ADA group through Week 72 (P ≤.01).Conclusion:The safety profile for UPA + MTX was consistent with that reported previously and with the integrated Phase 3 safety analysis.1,2UPA + MTX maintained significantly higher levels of clinical response, including remission compared to ADA + MTX through Week 72.References:[1]Fleischmann R, et al.Annals of the Rheumatic Diseases.2019;78:744-745.[2]Cohen SB, et al. Thu0167.Annals of the Rheumatic Diseases. 2019;78:357.Disclosure of Interests: :Roy Fleischmann Grant/research support from: AbbVie, Akros, Amgen, AstraZeneca, Bristol-Myers Squibb, Boehringer, IngelhCentrexion, Eli Lilly, EMD Serono, Genentech, Gilead, Janssen, Merck, Nektar, Novartis, Pfizer, Regeneron Pharmaceuticals, Inc., Roche, Samsung, Sandoz, Sanofi Genzyme, Selecta, Taiho, UCB, Consultant of: AbbVie, ACEA, Amgen, Bristol-Myers Squibb, Eli Lilly, Gilead, GlaxoSmithKline, Novartis, Pfizer, Sanofi Genzyme, UCB, In-Ho Song Shareholder of: AbbVie Inc., Employee of: AbbVie Inc., Jeffrey Enejosa Shareholder of: AbbVie Inc., Employee of: AbbVie Inc., Eduardo Mysler Grant/research support from: AbbVie, Amgen, Bristol Myers Squibb, Roche, Eli Lilly, Novartis, Janssen, Sanofi, and Pfizer., Speakers bureau: AbbVie, Amgen, Bristol Myers Squibb, Roche, Eli Lilly, Novartis, Janssen, Sanofi, and Pfizer, Louis Bessette Grant/research support from: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Eli Lilly, Janssen, Merck, Novartis, Pfizer, Roche, Sanofi, UCB Pharma, Consultant of: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Eli Lilly, Janssen, Merck, Novartis, Pfizer, Roche, Sanofi, UCB Pharma, Speakers bureau: AbbVie, Amgen, Bristol-Myers Squibb, Celgene, Eli Lilly, Janssen, Merck, Novartis, Pfizer, Sanofi, Patrick Durez Speakers bureau: AbbVie, Bristol-Myers Squibb, Celltrion, Eli Lilly, Pfizer, Sanofi, Andrew Ostor Consultant of: MSD, Pfizer, Lilly, Abbvie, Novartis, Roche, Gilead and BMS, Speakers bureau: MSD, Pfizer, Lilly, Abbvie, Novartis, Roche, Gilead and BMS, Jerzy Swierkot Grant/research support from: AbbVie, Sandoz, Pfizer, Roche, BMS, UCB, MSD, Accord, Janssen, Consultant of: AbbVie, Sandoz, Pfizer, Roche, BMS, UCB, MSD, Accord, Janssen, Speakers bureau: AbbVie, Sandoz, Pfizer, Roche, BMS, UCB, MSD, Accord, Janssen, Yanna Song Shareholder of: AbbVie Inc., Employee of: AbbVie Inc., Mark C. Genovese Grant/research support from: Abbvie, Eli Lilly and Company, EMD Merck Serono, Galapagos, Genentech/Roche, Gilead Sciences, Inc., GSK, Novartis, Pfizer Inc., RPharm, Sanofi Genzyme, Consultant of: Abbvie, Eli Lilly and Company, EMD Merck Serono, Genentech/Roche, Gilead Sciences, Inc., GSK, Novartis, RPharm, Sanofi Genzyme