Organisations that design and deploy artificial intelligence (AI) systems increasingly commit themselves to high-level, ethical principles. However, there still exists a gap between principles and ...practices in AI ethics. One major obstacle organisations face when attempting to operationalise AI Ethics is the lack of a well-defined material scope. Put differently, the question to which systems and processes AI ethics principles ought to apply remains unanswered. Of course, there exists no universally accepted definition of AI, and different systems pose different ethical challenges. Nevertheless, pragmatic problem-solving demands that things should be sorted so that their grouping will promote successful actions for some specific end. In this article, we review and compare previous attempts to classify AI systems for the purpose of implementing AI governance in practice. We find that attempts to classify AI systems proposed in previous literature use one of three mental models:
the Switch
, i.e., a binary approach according to which systems either are or are not considered AI systems depending on their characteristics;
the Ladder
, i.e., a risk-based approach that classifies systems according to the ethical risks they pose; and
the Matrix
, i.e., a multi-dimensional classification of systems that take various aspects into account, such as context, input data, and decision-model. Each of these models for classifying AI systems comes with its own set of strengths and weaknesses. By conceptualising different ways of classifying AI systems into simple mental models, we hope to provide organisations that design, deploy, or regulate AI systems with the vocabulary needed to demarcate the material scope of their AI governance frameworks.
While the use of artificial intelligence (AI) systems promises to bring significant economic and social benefits, it is also coupled with ethical, legal, and technical challenges. Business leaders ...thus face the question of how to best reap the benefits of automation whilst managing the associated risks. As a first step, many companies have committed themselves to various sets of ethics principles aimed at guiding the design and use of AI systems. So far so good. But how can well-intentioned ethical principles be translated into effective practice? And what challenges await companies that attempt to operationalize AI governance? In this article, we address these questions by drawing on our first-hand experience of shaping and driving the roll-out of AI governance within AstraZeneca, a biopharmaceutical company. The examples we discuss highlight challenges that any organization attempting to operationalize AI governance will have to face. These include questions concerning how to define the material scope of AI governance, how to harmonize standards across decentralized organizations, and how to measure the impact of specific AI governance initiatives. By showcasing how AstraZeneca managed these operational questions, we hope to provide project managers, CIOs, AI practitioners, and data privacy officers responsible for designing and implementing AI governance frameworks within other organizations with generalizable best practices. In essence, companies seeking to operationalize AI governance are encouraged to build on existing policies and governance structures, use pragmatic and action-oriented terminology, focus on risk management in development and procurement, and empower employees through continuous education and change management.
Purpose
Pharmacological actions of morphine are mediated through G‐protein‐coupled receptors (GPCRs), specifically mu‐opioid receptors (MORs). Morphine‐induced activation of GPCRs inhibits adenylyl ...cyclase and decreases 3’,5’‐cyclic adenosine monophosphate (cAMP). Neuroblastoma SH‐SY5Y cell lines endogenously express MORs. Our aim was to develop a flow cytometry (FC) assay for MORs on the surface of SH‐SY5Y cells. This assay will provide in‐depth understanding of opioid‐induced downstream cAMP signaling.
Methods
SH‐SY5Y human neuroblastoma cells (ATCC® CRL2266™) were grown in DMEM/10% FBS and maintained in a humidified incubator at 37ºC/5%CO2. Cells were seeded in 96‐well white, opaque‐bottom plates (5×104cells/well). Transfection conditions were optimized using ViaFect™ (Promega™) and a stably transfected SH‐SY5Y cell line was established using GloSensor™ ‐23F cAMP plasmid (Promega™) containing hygromycin‐resistance gene. Cells were washed in FACS buffer (PBS + 2% FBS) and incubated for 1h at room temperature with monoclonal rabbit anti‐MOR antibody (Ab) (Novus Biologicals). Following the wash, cells were incubated with donkey anti‐rabbit secondary Ab (Abcam). We used monoclonal mouse anti‐muscarinic Ab (Novus Biologicals) and donkey anti‐mouse secondary Ab (Jackson Labs). Both secondary Abs are bound to a fluorophore (Alexa Fluor® 488) which emits light when excited by FITC channel frequency (excitation/emission: 495nm/519nm). A Cytoflex flow cytometer (Beckman Coulter) analyzes cells for FITC fluorescence emission. Multiple primary and secondary concentrations were tested; Human TruStain Fc blocker (BioLegend) was used to prevent nonspecific tagging.
Results
Relative fluorescence emission was measured as percent shift increase from the untagged control group. Because MOR Abs have been tested in a limited capacity, we tested for presence of muscarinic receptors (another endogenously expressed GPCR in these cells) to establish the FC assay. Initial trials with low primary and secondary MOR Ab concentrations showed limited shifts (~1‐2%), however, muscarinic receptor Abs bound to their receptors at lower concentrations compared to MOR Abs. Testing muscarinic receptors showed secondary Ab concentration‐dependent fluorescence shifts indicating that the technique was viable. Studies with increased concentration ratios of primary and secondary MOR Abs showed secondary Ab concentration‐dependent response with maximum % shifts reaching up to 16%. Use of Fc blocker eliminated nonspecific binding, increasing the percent shift.
Conclusion
Compared to untagged controls, antibody‐tagging on SH‐SY5Y cell surface produced greater immunofluorescence indicating presence of MORs in both non‐transfected and 23F cAMP plasmid transfected cells. Direct binding of receptor modulators leads to altered downstream cAMP signaling as shown in previous in vitro signaling studies and behavioral mouse models. Detection of the MOR is essential to ensure that cAMP signaling is regulated by MOR agonists. In future studies, we plan to investigate the expression of endothelin‐A receptor (ETAR) and explore potential interactions between MOR and ETAR in SH‐SY5Y cells.
Background: COVID-19 vaccination can lead to herd immunity when a sufficient proportion of population is vaccinated. Effectiveness of vaccination coverage depends on the population’s willingness to ...be vaccinated completely. Aim and objective was to estimate COVID-19 vaccination coverage and to determine reasons for its acceptance and non-acceptance in urban field practice area of medical college attached with tertiary care hospital.
Methods: A cross-sectional house to house survey was conducted among the households (n=1346) under an urban health training centre (UHTC) of a medical college in central Gujarat after the second wave of COVID-19. All the beneficiaries above the age of 18 years residing in the house-holds coming under the field area were included in the study. The data was collected using Epi-collect 5 mobile based applications. Beneficiaries who had taken two doses of COVID-19 vaccine were counted as fully immunized. Vaccination coverage was calculated different for first and second dose. Facilitators and barriers for COVID-19 vaccination were shown with appropriate diagram.
Results: Out of the 1832 participants in the survey, vaccination coverage was 78.22% for first dose and 37.23% for second dose. Most common motivation (81%) behind taking vaccination was “good health”. Among the non-users for COVID-19 vaccination, 36.8% did not take the vaccine due to fear of adverse reactions.
Conclusions: Vaccination coverage is average in the urban area surveyed. Majority of the population is motivated to take the vaccine to protect themselves from COVID-19. The main reason for not taking vaccine was fear of side-effect of vaccine.
Organisations that design and deploy artificial intelligence (AI) systems increasingly commit themselves to high-level, ethical principles. However, there still exists a gap between principles and ...practices in AI ethics. One major obstacle organisations face when attempting to operationalise AI Ethics is the lack of a well-defined material scope. Put differently, the question to which systems and processes AI ethics principles ought to apply remains unanswered. Of course, there exists no universally accepted definition of AI, and different systems pose different ethical challenges. Nevertheless, pragmatic problem-solving demands that things should be sorted so that their grouping will promote successful actions for some specific end. In this article, we review and compare previous attempts to classify AI systems for the purpose of implementing AI governance in practice. We find that attempts to classify AI systems found in previous literature use one of three mental model. The Switch, i.e., a binary approach according to which systems either are or are not considered AI systems depending on their characteristics. The Ladder, i.e., a risk-based approach that classifies systems according to the ethical risks they pose. And the Matrix, i.e., a multi-dimensional classification of systems that take various aspects into account, such as context, data input, and decision-model. Each of these models for classifying AI systems comes with its own set of strengths and weaknesses. By conceptualising different ways of classifying AI systems into simple mental models, we hope to provide organisations that design, deploy, or regulate AI systems with the conceptual tools needed to operationalise AI governance in practice.
Recurrence rates after breast-conserving therapy may depend on genomic characteristics of cancer-adjacent, benign-appearing tissue. Studies have not evaluated recurrence in association with multiple ...genomic characteristics of cancer-adjacent breast tissue. To estimate the prevalence of DNA defects and RNA expression subtypes in cancer-adjacent, benign-appearing breast tissue at least 2 cm from the tumor margin, cancer-adjacent, pathologically well-characterized, benign-appearing breast tissue specimens from The Cancer Genome Atlas project were analyzed for DNA sequence, copy-number variation, DNA methylation, messenger RNA (mRNA) sequence, and mRNA/microRNA expression. Additional samples were also analyzed by at least one of these genomic data types and associations between genomic characteristics of normal tissue and overall survival were assessed. Approximately 40% of cancer-adjacent, benign-appearing tissues harbored genomic defects in DNA copy number, sequence, methylation, or in RNA sequence, although these defects did not significantly predict 10-year overall survival. Two mRNA/microRNA expression phenotypes were observed, including an active mRNA subtype that was identified in 40% of samples. Controlling for tumor characteristics and the presence of genomic defects, this active subtype was associated with significantly worse 10-year survival among estrogen receptor (ER)-positive cases. This multi-platform analysis of breast cancer-adjacent samples produced genomic findings consistent with current surgical margin guidelines, and provides evidence that extratumoral RNA expression patterns in cancer-adjacent tissue predict overall survival among patients with ER-positive disease.
While the use of artificial intelligence (AI) systems promises to bring significant economic and social benefits, it is also coupled with ethical, legal, and technical challenges. Business leaders ...thus face the question of how to best reap the benefits of automation whilst managing the associated risks. As a first step, many companies have committed themselves to various sets of ethics principles aimed at guiding the design and use of AI systems. So far so good. But how can well-intentioned ethical principles be translated into effective practice? And what challenges await companies that attempt to operationalize AI governance? In this article, we address these questions by drawing on our first-hand experience of shaping and driving the roll-out of AI governance within AstraZeneca, a biopharmaceutical company. The examples we discuss highlight challenges that any organization attempting to operationalize AI governance will have to face. These include questions concerning how to define the material scope of AI governance, how to harmonize standards across decentralized organizations, and how to measure the impact of specific AI governance initiatives. By showcasing how AstraZeneca managed these operational questions, we hope to provide project managers, CIOs, AI practitioners, and data privacy officers responsible for designing and implementing AI governance frameworks within other organizations with generalizable best practices. In essence, companies seeking to operationalize AI governance are encouraged to build on existing policies and governance structures, use pragmatic and action-oriented terminology, focus on risk management in development and procurement, and empower employees through continuous education and change management.
Study of feto-maternal outcome in twin pregnancy Sheth, Kenan; Sheikh, Amira M; Shah, Margi A
International Journal of Reproduction, Contraception, Obstetrics and Gynecology,
01/2023, Letnik:
12, Številka:
1
Journal Article
Odprti dostop
Background: In modern obstetrics with advanced ultrasonographic techniques and color Doppler, multiple pregnancy and associated condition like chorionicity, growth discordance, vascular ...complications, twin to twin transfusion syndrome, intrauterine death of one or more fetus and congenital anomalies can now be diagnosed at early stage of gestation. Vigilant obstetric care during antepartum, intrapartum and postpartum period decreases maternal morbidity and mortality as well as improves fetal outcome in twin pregnancy. Methods: This is a randomized prospective study of 250 patients of multiple pregnancy admitted at our institute from July 2020 to June 2022 including all emergency as well as registered patients. In all cases a detailed history was taken and all routine investigations were done. All information was entered in a proforma and the fetomaternal outcome of twin pregnancy was analysed. Results: Around 67% patients had onset of labor after 32 weeks of gestation, rest 32% patients had onset of labor at or before 32 weeks of gestation. The 55% patients underwent lower segment caesarean section and 45% patients delivered vaginally. We observed highest incidence of twins in age group 21-30 years. Maximum number of patients 59% were multigravida compared to primigravida. Conclusions: Multiple pregnancy is considered as "high risk pregnancy". hence early diagnosis of multiple pregnancy is essential in reducing maternal and perinatal morbidity and mortality. Keywords: Twins, Preterm, Perinatal morbidity, Perinatal mortality