Alba Franco

Bio: Alba Franco is an academic researcher. The author has contributed to research in topics: Population & Computer science. The author has co-authored 1 publications.

Analytical Performances of the COVISTIXTM Antigen Rapid Test for SARS-CoV-2 Detection in an Unselected Population (All-Comers)

Abstract: The performance and validity of the COVISTIXTM rapid antigen test for the detection of SARS-CoV-2 were evaluated in an unselected population. Additionally, we assessed the influence of the Omicron SARS-CoV-2 variant in the performance of this antigen rapid test. Swab samples were collected at two point-of-care facilities in Mexico City from individuals that were probable COVID-19 cases, as they were either symptomatic or asymptomatic persons at risk of infection due to close contact with SARS-CoV-2 positive cases. Detection of the Omicron SARS-CoV-2 variant was performed in 91 positive cases by Illumina sequencing. Specificity and sensitivity of the COVISTIXTM rapid antigen test was 96% (CI 95% 94–98) and 81% (CI 95% 76–85), respectively. The accuracy parameters were not affected in samples collected after 7 days of symptom onset, and it was possible to detect almost 65% of samples with a Ct-value between 30 and 34. The COVISTIXTM antigen rapid test is highly sensitive (93%; CI 95% 88–98) and specific (98%; CI 95% 97–99) for detecting Omicron SARS-CoV-2 variant carriers. The COVISTIXTM rapid antigen test is adequate for examining asymptomatic and symptomatic individuals, including those who have passed the peak of viral shedding, as well as carriers of the highly prevalent Omicron SARS-CoV-2 variant.

Analytical performances of the COVISTIX and Panbio antigen rapid tests for SARS-CoV-2 detection in an unselected population (all commers)

Abstract: Importance A steady increase in acute respiratory syndrome coronavirus 2 (SARS-CoV-2) cases worldwide is causing some regions of the world to withstand a third or even fourth wave of contagion. Swift detection of SARS-CoV-2 infection is paramount for the containment of cases, prevention of sustained contagion; and most importantly, for the reduction of mortality. Objective To evaluate the performance and validity of the COVISTIX™ rapid antigen test, for the detection of SARS-CoV-2 in an unselected population and compare it to Panbio™ rapid antigen test and RT-PCR. Design This is comparative effectiveness study; samples were collected at two point-of-care facilities in Mexico City between May and August 2021. Participants Recruited individuals were probable COVID-19 cases, either symptomatic or asymptomatic persons that were at risk of infection due to close contact to SARS-CoV-2 positive cases. Diagnostic intervention RT-PCR was used as gold standard for detection of SARS-CoV-2 in nasal and nasopharyngeal swabs, study subjects were tested in parallel either with the COVISTIX™ or with Panbio™ rapid antigen test. Main outcome Diagnostic performance of the COVISTIX™ assay is adequate in all commers since its accuracy parameters were not affected in samples collected after 7 days of symptom onset, and it detected almost 65% of samples with a Ct-value between 30 and 34. Results For the population tested with COVISTIX™ (n=783), specificity and sensitivity of the was 96.0% (CI95% 94.0-98.0) and 81% (CI95% 76.0-85.0), as for the Panbio™ (n=2202) population, was 99.0% (CI95%: 0.99-1.00) and 62% (CI%: 58.0-64.0%), respectively. Conclusions and relevance The COVISTIX™ rapid antigen test shows a high performance in all comers, thus, this test is also adequate for testing patients who have passed the peak of viral shedding or for asymptomatic patients.

AI-Assisted Thin Section Analysis, The Making of a Geologically-Realistic Digital Interpretation

Abstract: The visual interpretation of geological thin section is a meticulous endeavor carried out by geoscientific specialists in order to ground truth log interpretation as well as guide the spatial distribution of properties required by reservoir simulation models. At the same time, the shortage of qualified personnel, the abundance of dormant core data and the requirements for increased reservoir model accuracy have created operational needs that human interpreters alone can hardly fulfill. In this context, a method for AI-assisted thin section interpretation was developed, leveraging the latest advances in the field of deep learning to provide geologists with a comprehensive set of reservoir properties derived from rock images. While a significant part of the solution relies on the training of supervised convolutional neural networks, establishing consistent labeling procedure, enforcing geological rules, removing input and output image artifacts and close communication with subject matter experts were equally critical ingredients to a geologically-realistic prediction as well as supplementing a scarce amount of input training data. The main outcome of this multi-step domain-knowledge and data science work not only led to an increase in the mean intersection-of-union metric but also to the assurance that fundamental geological principles were honored. In practice, the algorithm ensured that petrographic object detection was constrained by biostatistical population criteria as well as prohibit the occurrence of non-natural combination of nested framework grain. The aforementioned enhancements were subsequentially implemented and deployed at company scale for ADNOC's specialists to carry out their geological interpretation through conventional web-browser applications.

Diagnosis of Melanocytic Lesions using the K-Nearest Neighbors classification technique

Abstract: Melanoma is a type of skin cancer that originates when melanocytes (the cells that give skin its tan or brown color) begin to grow out of control. Cells from almost any part of the body can develop into cancer and can then spread to other areas of the body. Currently there is no tool to assist the medical professional in the rapid diagnosis of melanoma. In this work, we propose a methodology that provides support to the dermatologist in the diagnosis of melanoma. This methodology is composed of 4 modules: Preprocessing, Segmentation, Feature Extraction and Classification. Image preprocessing consists of hair removal. Segmentation consists of isolating the object of interest (the lesion). The extracted features for classification are: asymmetry, border, color and dermoscopic structures. Finally, the K-nearest neighbor (K-NN) classification method is used to classify benign and malignant moles. The final results of the methodology show 91.67% accuracy, 92.50% sensitivity and 100% specificity. Since the results look promising, this technique could be the basis for more sophisticated tools useful to clinicians in the diagnosis of melanoma.