FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2992400503> ?p ?o ?g. }

• W2992400503 endingPage "e9" @default.
• W2992400503 startingPage "e8" @default.
• W2992400503 abstract "The advent of artificial intelligence (AI), big data, and the next-generation telecommunication network (5G) has generated enormous interest in digital health. Digital health comprises overlapping areas ranging from AI, the internet of things, electronic health, and telehealth to the analysis and use of big data.1WHOWHO guideline: recommendations on digital interventions for health system strengthening.https://www.who.int/reproductivehealth/publications/digital-interventions-health-system-strengthening/en/Date accessed: August 10, 2019Google Scholar With substantial innovation opportunities in digital health, WHO published a set of guidelines earlier this year,1WHOWHO guideline: recommendations on digital interventions for health system strengthening.https://www.who.int/reproductivehealth/publications/digital-interventions-health-system-strengthening/en/Date accessed: August 10, 2019Google Scholar advising potential researchers and innovators on how to harness this technology to create evidence-based interventions within real-world settings to improve patients' outcome. Here, we highlight some of the new developments in ophthalmology, focusing on AI and other digital innovation, particularly those that are clinically available and could be implemented in the foreseeable future. First, within the spectrum of digital health, AI and a subset of machine learning known as deep learning have become popular technologies and the forefront of the digital revolution. Compared with traditional feature-based machine learning techniques, deep learning has shown to outperform traditional techniques in image, speech, and motion recognition and natural language processing.2LeCun Y Bengio Y Hinton G Deep learning.Nature. 2015; 521: 436-444Crossref PubMed Scopus (41799) Google Scholar In medicine, deep learning has shown robust performance in detecting tuberculosis from chest x-rays, malignant melanoma on skin photographs, and lymph node metastases secondary to breast cancer from tissue sections. In ophthalmology, most of the deep learning systems focus heavily on image recognition, using fundus photographs and optical coherence tomography to detect diabetic retinopathy, diabetic macular oedema, glaucoma, age-related macular degeneration, retinopathy of prematurity, and cataract.3Ting DSW Peng L Varadarajan AV et al.Deep learning in ophthalmology: the technical and clinical considerations.Prog Retin Eye Res. 2019; 72100759Crossref PubMed Scopus (196) Google Scholar Excitingly, deep learning might be able to predict systemic cardiovascular risk factors such as age, sex, systemic blood pressure, and glycated haemoglobin A1c from the fundus photographs. Most of these new deep learning algorithms and systems, however, have been developed in retrospective, open-source, or cleaned research laboratory settings, with very few done in either real-world settings or from prospective studies. In different clinical areas in ophthalmology, the use of deep learning for diabetic retinopathy screening is the most established, the most commonly reported in the peer-reviewed literature, and the closest technology that could be translated, integrated, and implemented clinically. For example, in one of the first published prospective studies,4Gulshan V Rajan RP Widner K et al.Performance of a deep-learning algorithm vs manual grading for detecting diabetic retinopathy in India.JAMA Ophthalmol. 2019; 137: 987Crossref PubMed Scopus (96) Google Scholar deep learning achieved better diagnostic performance than the human graders in detecting referable diabetic retinopathy. The US Food and Drug Administration has approved the first medical AI-based device to screen for and monitor diabetic retinopathy in 2018,5Abràmoff MD Lavin PT Birch M Shah N Folk JC Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices.NPJ Digit Med. 2018; 1: 39Crossref PubMed Scopus (498) Google Scholar based on a preregistered prospective clinical trial that was done in primary eye-care settings in the USA. This approval has opened the door for introducing AI-based technology in the real-world setting. To date, about 30 million retinal optical coherence tomography procedures are now done every year in the USA for disease diagnosis and monitoring, which is more than all other ophthalmic imaging modalities combined. To tackle the burden of reading optical coherence tomography scans, an AI decision support tool might be a potential alternative. De Fauw and colleagues6De Fauw J Ledsam JR Romera-Paredes B et al.Clinically applicable deep learning for diagnosis and referral in retinal disease.Nat Med. 2018; 24: 1342-1350Crossref PubMed Scopus (1116) Google Scholar reported a novel deep learning system that can generate a referral triage decision, ensuring correct referral urgencies and siting of patients in the primary to tertiary eye-care settings. In the treatment monitoring of age-related macular degeneration, Schmidt-Erfurth and colleagues7Schmidt-Erfurth U Bogunovic H Sadeghipour A et al.Machine learning to analyze the prognostic value of current imaging biomarkers in neovascular age-related macular degeneration.Ophthalmol Retina. 2018; 2: 24-30Summary Full Text Full Text PDF PubMed Scopus (89) Google Scholar developed a machine learning model (random forest regression) that might predict visual acuity after 12 months in patients receiving ranibizumab for neovascular age-related macular degeneration, from the level of visual acuity at baseline, or within the first 3 months. This approach might help clinicians to counsel and better manage the patients' expectations appropriately during their treatment journey. Second, with AI, doctors can be excluded from the care pathway. Such a virtual clinic has been implemented into the eye hospital setting using the same principles—ie, when a patient has an eye scan with a specialist report either instead of or before a face-to-face consultation with an ophthalmologist. The scan reading, however, is only one component in the overall specialty care that these patients need. And so, if improvements are made in patient-reported outcomes, the use of an automated report as a substitute for the consultation needs to be carefully assessed to ascertain when it is appropriate to use fully automated services and when taking the direct interaction with the clinician out of the equation might reduce the quality of the holistic care the patient receives. Additionally, the internet of things is a new digital technology that can be used for AI deployment in ophthalmology.8Meinert E Van Velthoven M Brindley D et al.The internet of things in health care in Oxford: protocol for proof-of-concept projects.JMIR Res Protoc. 2018; 7e12077Crossref PubMed Scopus (20) Google Scholar With the internet of things, any devices connected to the internet can transfer information to each other in a purely automated way without the need for human–human or human–computer interaction. The application of the internet of things offers the opportunity to transform eye care by linking imaging data with clinical information not only within clinics, but between clinics in a region or cluster or even anywhere in the world. Images and clinical data can be analysed by AI algorithms or next-generation data analytics. One key example is the development of the digital virtual clinic, which, by negating the need for a physical appointment, addresses the major problem of patients having to wait for months for an eye appointment by highly trained specialists. In a UK proof-of-concept study using a cloud-based teleophthalmology platform that could be used across a range of devices, more than 50% of would-be referrals into hospital were prevented, allowing specialist care within local services in a cost-efficient way; although these findings need to be further validated clinically in the long term.9Kern C Fu DJ Kortuem K et al.Implementation of a cloud-based referral platform in ophthalmology: making telemedicine services a reality in eye care.Br J Ophthalmol. 2019; (bjophthalmol-2019-314161.)Crossref PubMed Scopus (40) Google Scholar The bidirectional communication of clinical information between ophthalmologists and optometrists further allowed embedding work-based learning into community optometry practices while upskilling local practitioners. In summary, health-care innovations, including AI and the internet of things, must start with the identification and the pull of unmet clinical needs, rather than the push of technology. Although studies on AI systems have been published, many of the successful AI algorithms are yet to be adopted within the clinical setting, including diabetic retinopathy screening algorithms. To expedite the translation of digital health into clinical care, it is important to gather, apart from the physical technology, all the key components of clinical care, including the overall health-care strategy, the standards and interoperability, infrastructure, legislation, policy, and compliance with clinical guidelines, and the appropriate workforce. This enabling framework is crucial to avoid the potential hype that a new disruptive innovation such as digital health could bring to the community,10Chen JH Asch SM Machine learning and prediction in medicine—beyond the peak of inflated expectations.N Engl J Med. 2017; 376: 2507-2509Crossref PubMed Scopus (471) Google Scholar with the proliferation of unconnected, short-lived, and poorly tested systems. DSWT and TYW are the co-inventors and patent holders of a deep learning system for retinal diseases. All other authors declare no competing interests." @default.
• W2992400503 created "2019-12-13" @default.
• W2992400503 creator A5032625133 @default.
• W2992400503 creator A5062880407 @default.
• W2992400503 creator A5072258594 @default.
• W2992400503 creator A5082967971 @default.
• W2992400503 creator A5088003995 @default.
• W2992400503 date "2020-01-01" @default.
• W2992400503 modified "2023-10-15" @default.
• W2992400503 title "Artificial intelligence, the internet of things, and virtual clinics: ophthalmology at the digital translation forefront" @default.
• W2992400503 cites W2621101817 @default.
• W2992400503 cites W2727650337 @default.
• W2992400503 cites W2886281300 @default.
• W2992400503 cites W2888424632 @default.
• W2992400503 cites W2896880582 @default.
• W2992400503 cites W2919115771 @default.
• W2992400503 cites W2942760134 @default.
• W2992400503 cites W2949074827 @default.
• W2992400503 doi "https://doi.org/10.1016/s2589-7500(19)30217-1" @default.
• W2992400503 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33328040" @default.
• W2992400503 hasPublicationYear "2020" @default.
• W2992400503 type Work @default.
• W2992400503 sameAs 2992400503 @default.
• W2992400503 citedByCount "47" @default.
• W2992400503 countsByYear W29924005032020 @default.
• W2992400503 countsByYear W29924005032021 @default.
• W2992400503 countsByYear W29924005032022 @default.
• W2992400503 countsByYear W29924005032023 @default.
• W2992400503 crossrefType "journal-article" @default.
• W2992400503 hasAuthorship W2992400503A5032625133 @default.
• W2992400503 hasAuthorship W2992400503A5062880407 @default.
• W2992400503 hasAuthorship W2992400503A5072258594 @default.
• W2992400503 hasAuthorship W2992400503A5082967971 @default.
• W2992400503 hasAuthorship W2992400503A5088003995 @default.
• W2992400503 hasBestOaLocation W29924005031 @default.
• W2992400503 hasConcept C104317684 @default.
• W2992400503 hasConcept C105580179 @default.
• W2992400503 hasConcept C110875604 @default.
• W2992400503 hasConcept C119767625 @default.
• W2992400503 hasConcept C136764020 @default.
• W2992400503 hasConcept C149364088 @default.
• W2992400503 hasConcept C41008148 @default.
• W2992400503 hasConcept C55493867 @default.
• W2992400503 hasConcept C71924100 @default.
• W2992400503 hasConcept C86803240 @default.
• W2992400503 hasConceptScore W2992400503C104317684 @default.
• W2992400503 hasConceptScore W2992400503C105580179 @default.
• W2992400503 hasConceptScore W2992400503C110875604 @default.
• W2992400503 hasConceptScore W2992400503C119767625 @default.
• W2992400503 hasConceptScore W2992400503C136764020 @default.
• W2992400503 hasConceptScore W2992400503C149364088 @default.
• W2992400503 hasConceptScore W2992400503C41008148 @default.
• W2992400503 hasConceptScore W2992400503C55493867 @default.
• W2992400503 hasConceptScore W2992400503C71924100 @default.
• W2992400503 hasConceptScore W2992400503C86803240 @default.
• W2992400503 hasIssue "1" @default.
• W2992400503 hasLocation W29924005031 @default.
• W2992400503 hasOpenAccess W2992400503 @default.
• W2992400503 hasPrimaryLocation W29924005031 @default.
• W2992400503 hasRelatedWork W1871685927 @default.
• W2992400503 hasRelatedWork W1985074889 @default.
• W2992400503 hasRelatedWork W2064446262 @default.
• W2992400503 hasRelatedWork W2353792954 @default.
• W2992400503 hasRelatedWork W2384747121 @default.
• W2992400503 hasRelatedWork W2509777158 @default.
• W2992400503 hasRelatedWork W2748952813 @default.
• W2992400503 hasRelatedWork W2899084033 @default.
• W2992400503 hasRelatedWork W4214508080 @default.
• W2992400503 hasRelatedWork W97426361 @default.
• W2992400503 hasVolume "2" @default.
• W2992400503 isParatext "false" @default.
• W2992400503 isRetracted "false" @default.
• W2992400503 magId "2992400503" @default.
• W2992400503 workType "article" @default.