A decision support system for the detection of cutaneous fungal infections using artificial intelligence.

Cutaneous fungal infections are one of the most common skin conditions, hence, the burden of determining fungal elements upon microscopic examination with periodic acid-Schiff (PAS) and Gomori methenamine silver (GMS) stains, is very time consuming. Despite some morphological variability posing challenges to training artificial intelligence (AI)-based solutions, these structures are favored potential targets, enabling the recruitment of promising AI-based technologies. Herein, we present a novel AI solution for identifying skin fungal infections, potentially providing a decision support system for pathologists. Skin biopsies of patients diagnosed with a cutaneous fungal infection at the Sheba Medical Center, Israel between 2014 and 2023, were used. Samples were stained with PAS and GMS and digitized by the Philips IntelliSite scanner. DeePathology® STUDIO fungal elements were annotated and deemed as ground truth data after an overall revision by two specialist pathologists. Subsequently, they were used to create an AI-based solution, which has been further validated in other regions of interests. The study participants were divided into two cohorts. In the first cohort, the overall sensitivity of the algorithm was 0.8, specificity 0.97, F1 score 0.78; in the second, the overall sensitivity of the algorithm was 0.93, specificity 0.99, F1 score 0.95. The results obtained are encouraging as proof of concept for an AI-based fungi detection algorithm. DeePathology® STUDIO can be employed as a decision support system for pathologists when diagnosing a cutaneous fungal infection using PAS and GMS stains, thereby, saving time and money.

Reliability of blood tests taken from the peripheral intravenous catheter.

We aimed to compare the reliability of laboratory blood tests using 2 sampling methods, via a peripheral venous catheter (PVC) vs direct venipuncture stab (DVS), we evaluated the effect of time elapsed since PVC insertion, PVC diameter, and administration of saline and/or antibiotic infusion through PVC on the blood test results. A prospective comparative study was conducted between May 2018 and July 2019. Patients aged ≥ 18 years and admitted to our department with a 20G/22G PVC inserted within the last 24 hours were enrolled. Blood samples were collected from each participant in the morning, and a second sample was drawn using PVC. Dependent variables included the percentage of hemolysis, failure rate, complete blood count, biochemical testing parameters, and coagulation functions. A total of 211 patients participated in the study. In total, 237 blood tests were conducted, of which 167 were performed on day 1 and the remaining on day 2, with a second blood sample collected from 26 patients on day 2. Twenty-one participants received 22G PVC, and 23 participants received active infusion. No significant differences were found in failure rates when each subgroup was compared with the primary day 1 group. The intraclass correlation coefficient indicated significant correlations among all the indices in all groups. Both blood sampling methods (PVC and direct venipuncture) can be used interchangeably for routine laboratory tests on days 1 and 2 after PVC insertion using 20G/22G PVC or infused PVC.

The urologist's learning curve of "in-bore" magnetic resonance-guided prostate biopsy.

BACKGROUND: The combination of multi-parametric MRI to locate and define suspected lesions together with their being targeted by an MRI-guided prostate biopsy has succeeded in increasing the detection rate of clinically significant disease and lowering the detection rate of non-significant prostate cancer. In this work we investigate the urologist's learning curve of in-bore MRI-guided prostate biopsy which is considered to be a superior biopsy technique. MATERIALS AND METHODS: Following Helsinki approval by The Chaim Sheba Medical Center ethics committee in accordance with The Sheba Medical Center institutional guidelines (5366-28-SMC) we retrospectively reviewed 110 IB-MRGpBs performed from 6/2016 to 1/2019 in a single tertiary center. All patients had a prostate multi-parametric MRI finding of at least 1 target lesion (prostate imaging reporting and data system [PI-RADS] score ≥ 3). We analyzed biopsy duration and clinically significant prostate cancer detection of targeted sampling in 2 groups of 55 patients each, once by a urologist highly trained in IB-MRGpBs and again by a urologist untrained in IB-MRGpBs. These two parameters were compared according to operating urologist and chronologic order. RESULTS: The patients' median age was 68 years (interquartile range 62-72). The mean prostate-specific antigen level and prostate size were 8.6 ± 9.1 ng/d and 53 ± 27 cc, respectively. The mean number of target lesions was 1.47 ± 0.6. Baseline parameters did not differ significantly between the 2 urologists' cohorts. Overall detection rates of clinically significant prostate cancer were 19%, 55%, and 69% for PI-RADS 3, 4 and 5, respectively. Clinically significant cancer detection rates did not differ significantly along the timeline or between the 2 urologists. The average duration of IB-MRGpB targeted sampling was 28 ± 15.8 min, correlating with the number of target lesions (p < 0.0001), and independent of the urologist's expertise. Eighteen cases defined the cutoff for the procedure duration learning curve (p < 0.05). CONCLUSIONS: Our data suggest a very short learning curve for IB-MRGpB-targeted sampling duration, and that clinically significant cancer detection rates are not influenced by the learning curve of this technique.

Late onset sepsis: comparison between coagulase-negative staphylococci and other bacteria in the neonatal intensive care unit.

OBJECTIVE: To compare demographic and clinical features of neonates with late-onset sepsis due to coagulase-negative-staphylococcus with those due to other bacterial pathogens. STUDY DESIGN: Retrospective, population-based cohort study. Data on infants less than 90 days old diagnosed with late-onset bacterial sepsis in the neonatal intensive care unit were reviewed. Univariable and multivariable analysis were performed. RESULTS: Two hundred and sixteen sepsis episodes were identified: coagulase-negative-staphylococcus caused 113 (52.3%) and other pathogen 103 (47.7%). Patients with coagulase-negative-staphylococcus sepsis had lower gestational age and younger age at onset, higher rates of parenteral nutrition exposure and normal temperature, lower rates of necrotizing enterocolitis, meningitis and neutropenia than patients with sepsis caused by other pathogens. In multivariable analysis, parenteral nutrition was the only independent risk factor for coagulase-negative-staphylococcus sepsis (odds ratio: 3.5, 95% confidence interval: 1.4-8.6). CONCLUSIONS: Initial empiric treatment for suspected sepsis should be targeted for other pathogens than coagulase-negative-staphylococci and vancomycin treatment should be reserved for infants with specific risk factors and according to local antimicrobial susceptibility.