FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4205691169> ?p ?o ?g. }

• W4205691169 endingPage "e135" @default.
• W4205691169 startingPage "e135" @default.
• W4205691169 abstract "Barack Obama once said that Black History Month was about “taking an unvarnished look at the past so we can create a better future”. Ideally, artificial intelligence (AI) in medicine should do just that, by analysing data from the past to improve health-care outcomes of patients in the future. However, experts have warned that AI models can instead replicate and amplify systematic racism. How can we better leverage AI to create racial equality in health? In The Lancet Digital Health, McCradden and colleagues say that the true potential of AI lies in revealing existing biases to motivate societal change and correct disparities in health care. For example, the disparity in pain management between Black and White patients is systemic, driven by biases with no scientific basis; however, race and ethnicity are still used as a proxy for indicators of health. Pierson and colleagues developed an AI model to measure osteoarthritis severity linked to knee pain, and found that AI predictions correlated more closely with patient-reported pain than with radiologists' diagnosis, particularly in Black patients. The study, published in Nature Medicine, highlighted the algorithm's ability to reduce unexplained disparities entrenched in the racial and socioeconomic diversity of the data. The authors state that the AI may better capture under-served patients' pain and potentially redress disparities in access to further treatments. Although examples of AI that expose disparity in health care are growing, these studies are still rare. Besides, the historical omission of medical data from minority populations and inequities in social and structural determinants of health cannot be undone by an algorithm. Contributing to these shortcomings might be the perilous state of diversity across the technology sector. In Nature Medicine, Chen and colleagues explain that increasing the diversity of our workforce is essential to treating health disparities with AI, stating that “more diverse perspectives will ensure that the right questions are asked”. Addressing the workforce diversity dilemma in the USA, President Joe Biden has stood by his pledge to combat inequality and racism in health and society in his appointment of Alondra Nelson as Deputy Director for Science and Society. Nelson has studied the impacts of technology and racism in science and medicine. In particular, her work has helped reveal the role of genetics in understanding racial health inequality. Her appointment comes at a critical time, as COVID-19 continues to disproportionately increase morbidity and mortality of Black people in the USA, despite reports that Black populations in the USA are more likely to engage in risk-reducing behaviours, such as mask-wearing. Although Nelson's appointment to the White House has introduced much-needed diversity in medicine and technology, more needs to be done to promote and retain Black researchers into senior positions. The recent dismissal of a prominent Google AI and ethics researcher and co-founder of Black in AI, Timnit Gebru, has sent shockwaves through the AI community in the USA. It has been widely reported that the cause of her departure may be linked to Gebru's published research findings, which are critical of Google's AI technology. More than 2600 Google employees and 4300 academic, industry, and civil society supporters have signed a public letter that says Gebru's departure “heralds danger for people working for ethical and just AI—especially Black people and People of Color—across Google”. As a leader in the field of AI and health care, Google has a responsibility to set standards for AI research. All research must meet global ethical standards of integrity and AI researchers must understand the diverse needs of those reliant on their algorithms. Likewise, within the research community, Knight and colleagues describe the essential role of data scientists in tackling racism and discrimination in health care, and emphasise that generalisability, transparency, and reproducibility are essential for ethical and race-sensitive data-driven insights. AI has the potential to help create racial equality in health. However, increasing diversity in the workforce that creates health-care algorithms is critical to developing AI that can overcome health-care disparities. Companies and research institutes must invest in women and people of colour to ensure the next generation of AI in medicine serves us all, equally." @default.
• W4205691169 created "2022-01-26" @default.
• W4205691169 creator A5064751349 @default.
• W4205691169 date "2021-03-01" @default.
• W4205691169 modified "2023-09-23" @default.
• W4205691169 title "Can artificial intelligence help create racial equality in the USA?" @default.
• W4205691169 doi "https://doi.org/10.1016/s2589-7500(21)00023-6" @default.
• W4205691169 hasPublicationYear "2021" @default.
• W4205691169 type Work @default.
• W4205691169 citedByCount "3" @default.
• W4205691169 countsByYear W42056911692021 @default.
• W4205691169 countsByYear W42056911692022 @default.
• W4205691169 countsByYear W42056911692023 @default.
• W4205691169 crossrefType "journal-article" @default.
• W4205691169 hasAuthorship W4205691169A5064751349 @default.
• W4205691169 hasBestOaLocation W42056911691 @default.
• W4205691169 hasConcept C139838865 @default.
• W4205691169 hasConcept C15744967 @default.
• W4205691169 hasConcept C160735492 @default.
• W4205691169 hasConcept C17744445 @default.
• W4205691169 hasConcept C199539241 @default.
• W4205691169 hasConcept C2250968 @default.
• W4205691169 hasConcept C2776515129 @default.
• W4205691169 hasConcept C71924100 @default.
• W4205691169 hasConcept C74909509 @default.
• W4205691169 hasConceptScore W4205691169C139838865 @default.
• W4205691169 hasConceptScore W4205691169C15744967 @default.
• W4205691169 hasConceptScore W4205691169C160735492 @default.
• W4205691169 hasConceptScore W4205691169C17744445 @default.
• W4205691169 hasConceptScore W4205691169C199539241 @default.
• W4205691169 hasConceptScore W4205691169C2250968 @default.
• W4205691169 hasConceptScore W4205691169C2776515129 @default.
• W4205691169 hasConceptScore W4205691169C71924100 @default.
• W4205691169 hasConceptScore W4205691169C74909509 @default.
• W4205691169 hasIssue "3" @default.
• W4205691169 hasLocation W42056911691 @default.
• W4205691169 hasOpenAccess W4205691169 @default.
• W4205691169 hasPrimaryLocation W42056911691 @default.
• W4205691169 hasRelatedWork W2020606322 @default.
• W4205691169 hasRelatedWork W2499897038 @default.
• W4205691169 hasRelatedWork W2748952813 @default.
• W4205691169 hasRelatedWork W2899084033 @default.
• W4205691169 hasRelatedWork W3000437961 @default.
• W4205691169 hasRelatedWork W3116924934 @default.
• W4205691169 hasRelatedWork W3198286774 @default.
• W4205691169 hasRelatedWork W3210792304 @default.
• W4205691169 hasRelatedWork W4253921463 @default.
• W4205691169 hasRelatedWork W4297990487 @default.
• W4205691169 hasVolume "3" @default.
• W4205691169 isParatext "false" @default.
• W4205691169 isRetracted "false" @default.
• W4205691169 workType "article" @default.