• thina

Umbono waseCanada wokufundisa ubuhlakani bokwenziwa kubafundi bezokwelapha

Siyabonga ngokuvakashela i-Nature.com.Inguqulo yesiphequluli oyisebenzisayo inosekelo olukhawulelwe lwe-CSS.Ukuze uthole imiphumela engcono kakhulu, sincoma ukuthi usebenzise inguqulo entsha yesiphequluli sakho (noma uvale imodi yokuvumelana ku-Internet Explorer).Okwamanje, ukuze siqinisekise ukwesekwa okuqhubekayo, sibonisa isayithi ngaphandle kwesitayela noma i-JavaScript.
Izicelo ze-clinical artificial intelligence (AI) zikhula ngokushesha, kodwa izifundo ezikhona zesikole sezokwelapha zinikeza ukufundisa okulinganiselwe okuhlanganisa le ndawo.Lapha sichaza isifundo sokuqeqeshwa kobuhlakani bokwenziwa esikusungule futhi sakulethwa kubafundi bezokwelapha baseCanada futhi senza izincomo zokuqeqeshwa okuzayo.
I-Artificial Intelligence (AI) kwezokwelapha ingathuthukisa ukusebenza kahle kwendawo yokusebenza futhi isize ekuthathweni kwezinqumo zomtholampilo.Ukuze baqondise ngokuphepha ukusetshenziswa kobuhlakani bokwenziwa, odokotela kufanele babe nokuqonda okuthile ngobuhlakani bokwenziwa.Amazwana amaningi akhuthaza ukufundisa imiqondo ye-AI1, njengokuchaza amamodeli we-AI nezinqubo zokuqinisekisa2.Kodwa-ke, zimbalwa izinhlelo ezihleliwe eseziqalisiwe, ikakhulukazi ezingeni likazwelonke.Pinto dos Santos et al.3.Abafundi bezobudokotela abangu-263 bahlolwa kwathi u-71% bavuma ukuthi badinga ukuqeqeshwa kwezobunhloli bokwenziwa.Ukufundisa ubuhlakani bokwenziwa ezithamelini zezokwelapha kudinga idizayini ecophelelayo ehlanganisa imiqondo yezobuchwepheshe nengeyona yezobuchwepheshe kubafundi abavame ukuba nolwazi oluningi lwangaphambili.Sichaza ulwazi lwethu ngokuletha uchungechunge lwemihlangano yokufundisa ye-AI emaqenjini amathathu abafundi bezokwelapha futhi senze izincomo zemfundo yezokwelapha yesikhathi esizayo ku-AI.
Isethulo sethu samaviki amahlanu se-Artificial Intelligence in Medicine workshop yabafundi bezokwelapha sabanjwa izikhathi ezintathu phakathi kukaFebhuwari 2019 no-April 2021. Ishejuli yeshabhu ngayinye yokusebenzela, enencazelo emfushane yezinguquko zesifundo, iboniswa kuMfanekiso 1. Isifundo sethu izinjongo zokufunda ezintathu eziyinhloko: abafundi bayaqonda ukuthi idatha icutshungulwa kanjani ekusetshenzisweni kobuhlakani bokwenziwa, bahlaziye izincwadi zobuhlakani bokwenziwa zezicelo zomtholampilo, futhi basebenzise amathuba okuhlanganyela nonjiniyela abathuthukisa ubuhlakani bokwenziwa.
Okuluhlaza okwesibhakabhaka isihloko senkulumo kanye nokukhanya okuluhlaza okwesibhakabhaka isikhathi sokuphendulana sombuzo kanye nezimpendulo.Ingxenye empunga igxile ekubuyekezweni okufushane kwezincwadi.Izigaba eziwolintshi ziyizibonelo ezikhethiwe ezichaza amamodeli noma amasu obuhlakani bokwenziwa.Okuluhlaza kuyisifundo sokuhlela esiqondisiwe esiklanyelwe ukufundisa ubuhlakani bokwenziwa ukuxazulula izinkinga zomtholampilo nokuhlola amamodeli.Okuqukethwe kanye nobude besikhathi sama-workshops kuyehluka ngokuya ngokuhlolwa kwezidingo zabafundi.
Iworkshop yokuqala yaba seNyuvesi yaseBritish Columbia kusukela ngoFebhuwari kuya ku-Ephreli 2019, futhi bonke ababambiqhaza abayisi-8 banikeze impendulo eyakhayo4.Ngenxa ye-COVID-19, umhlangano wokucobelelana ngolwazi wesibili wabanjwa cishe ngo-Okthoba-Novemba 2020, nabafundi bezokwelapha abangu-222 kanye nezakhamuzi ezi-3 ezivela ezikoleni zezokwelapha zaseCanada eziyisi-8.Amaslayidi ephrezentheshini nekhodi kulayishwe endaweni evulekile yokufinyelela (http://ubcaimed.github.io).Impendulo eyinhloko evela ekuphindaphindweni kokuqala yayiwukuthi izinkulumo beziqine kakhulu futhi izinto ezisetshenziswayo ziwuthiyori kakhulu.Ukukhonza izindawo zezikhathi eziyisithupha zaseCanada kubangela izinselele ezengeziwe.Ngakho-ke, iworkshop yesibili yafushanisa iseshini ngayinye yaba yihora elingu-1, yenza izinto zesifundo zaba lula, yengeza izifundo eziyizibonelo ezengeziwe, futhi yakha izinhlelo ze-boilerplate ezivumela ababambiqhaza ukuthi baqedele amazwibela ekhodi ngokususa amaphutha okuncane (Ibhokisi 1).Impendulo engukhiye evela ekuphindaphindweni kwesibili ihlanganise impendulo eyakhayo ekuzilolongeni kokuhlela kanye nesicelo sokubonisa ukuhlela iphrojekthi yokufunda komshini.Ngakho-ke, kuworkshop yethu yesithathu, ebibanjwe cishe izitshudeni zezokwelapha eziyi-126 ngoMashi-April 2021, sifake izivivinyo ezibandakanyayo zokubhala amakhodi namaseshini empendulo yephrojekthi ukuze sibonise umthelela wokusebenzisa imiqondo yeshabhu kumaphrojekthi.
Ukuhlaziywa Kwedatha: Inkambu yocwaningo kwizibalo ekhomba amaphethini abalulekile kudatha ngokuhlaziya, ukucubungula, kanye namaphethini edatha.
Ukumbiwa kwedatha: inqubo yokuhlonza kanye nokukhipha idatha.Kumongo wobuhlakani bokwenziwa, lokhu kuvame ukuba kukhulu, nokuhlukahluka okuningi kwesampula ngayinye.
Ukuncishiswa kobukhulu: Inqubo yokuguqula idatha enezici eziningi ezingazodwana zibe izici ezimbalwa kuyilapho kugcinwa izici ezibalulekile zesethi yedatha yoqobo.
Izici (kokuqukethwe kobuhlakani bokwenziwa): izakhiwo ezilinganisekayo zesampula.Ngokuvamile isetshenziswa ngokushintshana ngokuthi “impahla” noma “okuguquguqukayo”.
Imephu Yokuvula I-Gradient: Indlela esetshenziswa ukuhumusha amamodeli obuhlakani bokwenziwa (ikakhulukazi amanethiwekhi e-convolutional neural), ehlaziya inqubo yokuthuthukisa ingxenye yokugcina yenethiwekhi ukuze kukhonjwe izifunda zedatha noma izithombe eziqagela kakhulu.
Imodeli Ejwayelekile: Imodeli ekhona ye-AI eqeqeshwe kusengaphambili ukwenza imisebenzi efanayo.
Ukuhlola (kumongo wobuhlakani bokwenziwa): ukubheka ukuthi imodeli yenza kanjani umsebenzi isebenzisa idatha engakaze ihlangabezane nayo ngaphambilini.
Ukuqeqeshwa (kumongo wobuhlakani bokwenziwa): Ukunikeza imodeli ngedatha nemiphumela ukuze imodeli ilungise amapharamitha ayo angaphakathi ukuze ithuthukise amandla ayo okwenza imisebenzi isebenzisa idatha entsha.
I-Vector: uhlu lwedatha.Ekufundeni komshini, i-elementi ngayinye yamalungu afanayo ngokuvamile iyisici esihlukile sesampula.
Ithebula 1 libala izifundo zakamuva zango-Ephreli 2021, okuhlanganisa nezinhloso zokufunda ezihlosiwe zesihloko ngasinye.Le workshop ihloselwe labo abasanda kufika ezingeni lobuchwepheshe futhi ayidingi noma yiluphi ulwazi lwezibalo ngale konyaka wokuqala weziqu zobudokotela.Lesi sifundo sithuthukiswe ngabafundi bezokwelapha abayisi-6 nothisha abathathu abaneziqu eziphakeme kwezobunjiniyela.Onjiniyela bathuthukisa ithiyori yobuhlakani bokwenziwa ukuze bayifundise, futhi abafundi bezokwelapha bafunda izinto eziphathelene nomtholampilo.
Ama-workshops ahlanganisa izinkulumo, izifundo zezenzakalo, kanye nezinhlelo eziqondisiwe.Enkulumweni yokuqala, sibuyekeza imiqondo ekhethiwe yokuhlaziywa kwedatha ku-biostatistics, okuhlanganisa ukubonwa kwedatha, ukuhlehla kwezinto, kanye nokuqhathaniswa kwezibalo ezichazayo nezifundisayo.Nakuba ukuhlaziywa kwedatha kuyisisekelo sobuhlakani bokwenziwa, asizibandakanyi izihloko ezifana nokumbiwa kwedatha, ukuhlola ukubaluleka, noma ukuboniswa okusebenzisanayo.Lokhu kwakungenxa yemikhawulo yesikhathi futhi nangenxa yokuthi abanye abafundi abathola iziqu babenokuqeqeshwa kwangaphambili kwe-biostatistics futhi babefuna ukuhlanganisa izihloko ezihlukile zokufunda ngomshini.Inkulumo elandelayo yethula izindlela zesimanje futhi idingida ukwakheka kwezinkinga ze-AI, izinzuzo kanye nemikhawulo yamamodeli e-AI, nokuhlolwa kwamamodeli.Izinkulumo ziphelezelwa izincwadi kanye nocwaningo olusebenzayo mayelana nemishini yezobunhloli ekhona.Sigcizelela amakhono adingekayo ukuze kuhlolwe ukusebenza kahle nokuba nokwenzeka kwemodeli ukuze kubhekwane nemibuzo yomtholampilo, okuhlanganisa nokuqonda imikhawulo yamadivayisi obuhlakani bokwenziwa akhona.Isibonelo, sicele abafundi ukuthi bahumushe imihlahlandlela yokulimala kwekhanda yezingane ehlongozwe u-Kupperman et al., 5 esebenzisa i-algorithm yomuthi wesinqumo sobuhlakani bokwenziwa ukuze banqume ukuthi ingabe i-CT scan ingaba usizo yini ngokusekelwe ekuhlolweni kukadokotela.Sigcizelela ukuthi lesi yisibonelo esivamile se-AI ehlinzeka ngezibalo zokubikezela ukuze odokotela batolike, kunokuba bathathele odokotela esikhundleni.
Ezibonelweni ezitholakalayo zohlelo lwe-bootstrap yomthombo ovulekile (https://github.com/ubcaimed/ubcaimed.github.io/tree/master/programming_examples), sibonisa indlela yokuhlaziya idatha yokuhlola, ukunciphisa ubukhulu, ukulayisha imodeli evamile, nokuqeqeshwa .kanye nokuhlola.Sisebenzisa ama-notebook e-Google Collaboratory (Google LLC, Mountain View, CA), avumela ikhodi ye-Python ukuthi isetshenziswe kusiphequluli sewebhu.Emfanekisweni 2 unikeza isibonelo somsebenzi wokuhlela.Lo msebenzi ubandakanya ukubikezela izifo zisebenzisa i-Wisconsin Open Breast Imaging Dataset6 kanye ne-algorithm yesihlahla sesinqumo.
Yethula izinhlelo isonto lonke ngezihloko ezihlobene futhi ukhethe izibonelo ezinhlelweni ezishicilelwe ze-AI.Izakhi zokuhlela zifakwa kuphela uma zibhekwa njengezibalulekile ekunikezeni ukuqonda ekusebenzeni komtholampilo kwesikhathi esizayo, njengokuthi ungawahlola kanjani amamodeli ukuze kunqunywe ukuthi akulungele yini ukusetshenziswa ezivivinyweni zomtholampilo.Lezi zibonelo zifinyelela umvuthwandaba ngohlelo lokusebenza olugcwele oluphelele oluhlukanisa izimila njengeziyingozi noma eziyingozi ngokusekelwe kumapharamitha wesithombe sezokwelapha.
I-Heterogeneity yolwazi lwangaphambili.Abahlanganyeli bethu bahluka ngezinga labo lolwazi lwezibalo.Isibonelo, abafundi abanezizinda zobunjiniyela ezithuthukile bafuna izinto ezijulile, njengokuthi bazenza kanjani ezabo izinguquko ze-Fourier.Nokho, ukuxoxa nge-algorithm ye-Fourier ekilasini akwenzeki ngoba kudinga ulwazi olujulile lokucubungula isignali.
Ukuphuma kwezihambeli.Ukuba khona emihlanganweni yokulandelela kunqatshiwe, ikakhulukazi ngamafomethi aku-inthanethi.Isixazululo kungaba ukulandelela abakhona kanye nokunikeza isitifiketi sokuqedwa.Izikole zezokwelapha zaziwa ngokubona imibhalo yemisebenzi yabafundi yangaphandle kwesikole, engakhuthaza abafundi ukuthi baphishekele iziqu.
Idizayini Yesifundo: Ngenxa yokuthi i-AI ihlanganisa izinkundla eziningi ezingaphansi, ukukhetha imiqondo ewumongo yokujula nobubanzi obufanele kungaba inselele.Isibonelo, ukuqhubeka kokusetshenziswa kwamathuluzi e-AI kusukela elabhorethri kuya emtholampilo kuyisihloko esibalulekile.Nakuba sihlanganisa ukucutshungulwa kwangaphambili kwedatha, ukwakhiwa kwamamodeli, nokuqinisekisa, asifaki izihloko ezifana nokuhlaziywa kwedatha enkulu, ukubona ngeso lengqondo, noma ukwenza izivivinyo zomtholampilo ze-AI, kunalokho sigxila emicabangweni ye-AI ehluke kakhulu.Umgomo wethu osiqondisayo uwukuthuthukisa ukufunda nokubhala, hhayi amakhono.Isibonelo, ukuqonda ukuthi imodeli icubungula kanjani izici zokufakwayo kubalulekile ekuchazekeni.Enye indlela yokwenza lokhu ukusebenzisa amamephu okwenza kusebenze i-gradient, angabona ngeso lengqondo ukuthi yiziphi izifunda zedatha ezingabikezelwa.Nokho, lokhu kudinga i-multivariate calculus futhi ayikwazi ukwethulwa8.Ukwakhiwa kwamagama ajwayelekile bekuyinselele ngoba besizama ukuchaza ukuthi kusetshenzwa kanjani ngedatha njengama-vector ngaphandle kokusemthethweni kwezibalo.Qaphela ukuthi amagama ahlukene anencazelo efanayo, isibonelo, ku-epidemiology, "isici" sichazwa ngokuthi "okuguquguqukayo" noma "isibaluli."
Ukugcinwa kolwazi.Ngenxa yokuthi ukusetshenziswa kwe-AI kunomkhawulo, izinga ababambiqhaza abagcina ngalo ulwazi kusazobonakala.Ikharikhulamu yesikole sezokwelapha ivamise ukuncika ekuphindaphindweni kwezikhala ukuze kuqiniswe ulwazi ngesikhathi sokuzungezisa okusebenzayo,9 okungase futhi kusetshenziswe emfundweni ye-AI.
Ubungcweti bubaluleke kakhulu kunokwazi ukufunda nokubhala.Ukujula kokuqukethwe kuklanywe ngaphandle kokuqina kwezibalo, obekuyinkinga ngenkathi kwethulwa izifundo zomtholampilo zobuhlakani bokwenziwa.Ezibonelweni zohlelo, sisebenzisa uhlelo lwesifanekiso oluvumela ababambiqhaza ukuthi bagcwalise izinkambu futhi baqhube isofthiwe ngaphandle kokuthola ukuthi ungasetha kanjani indawo ephelele yokuhlela.
Ukukhathazeka mayelana nobuhlakani bokwenziwa okukhulunywa ngakho: Kunokukhathazeka okusabalele ukuthi ubuhlakani bokwenziwa bungase buthathe indawo yemisebenzi ethile yomtholampilo3.Ukuze sibhekane nalolu daba, sichaza imikhawulo ye-AI, okuhlanganisa neqiniso lokuthi cishe bonke ubuchwepheshe be-AI obugunyazwe abalawuli budinga ukugadwa kodokotela11.Siphinde sigcizelele ukubaluleka kokuchema ngoba ama-algorithms athambekele ekuchemani, ikakhulukazi uma isethi yedatha ingehlukene12.Ngenxa yalokho, iqembu elincane elithile lingase limodelwe ngendlela engalungile, okuholela ezinqumweni zomtholampilo ezingalungile.
Izinsiza zitholakala esidlangalaleni: Sidale izinsiza ezitholakala esidlangalaleni, okuhlanganisa amaslayidi ezinkulumo namakhodi.Nakuba ukufinyelela kokuqukethwe okuvumelanayo kunqunyelwe ngenxa yezindawo zesikhathi, okuqukethwe komthombo ovulekile kuyindlela elula yokufunda ehambisanayo njengoba ubuchwepheshe be-AI bungatholakali kuzo zonke izikole zezokwelapha.
Ukusebenzisana Kwemikhakha Ehlukahlukene: Lo mhlangano wokucobelelana ngolwazi uyibhizinisi elihlanganyelwe eliqalwe ngabafundi bezokwelapha ukuhlela izifundo kanye nonjiniyela.Lokhu kukhombisa amathuba okubambisana kanye negebe lolwazi kuzo zombili lezi zindawo, okuvumela ababambiqhaza ukuthi baqonde indima engaba khona abangayifaka isandla esikhathini esizayo.
Chaza amakhono asemqoka we-AI.Ukuchaza uhlu lwamakhono kunikeza isakhiwo esimisiwe esingahlanganiswa nekharikhulamu yezokwelapha esekelwe emakhono akhona.Le workshop njengamanje isebenzisa i-Learning Objective Levels 2 (Yokuqonda), 3 (Isicelo), kanye ne-4 (Ukuhlaziya) ye-Bloom's Taxonomy.Ukuba nezinsiza emazingeni aphezulu okuhlukanisa, njengokwenza amaphrojekthi, kungaqinisa ulwazi.Lokhu kudinga ukusebenzisana nochwepheshe bemitholampilo ukuze kunqunywe ukuthi izihloko ze-AI zingasetshenziswa kanjani ekugelezeni komsebenzi wezokwelapha kanye nokuvimbela ukufundiswa kwezihloko eziphindaphindwayo esezifakiwe ezinhlelweni zezifundo zezokwelapha ezijwayelekile.
Dala izibonelo zezifundo usebenzisa i-AI.Ngokufanayo nezibonelo zemitholampilo, ukufunda okusekelwe ecaleni kungaqinisa imiqondo engabonakali ngokugqamisa ukuhlobana kwayo emibuzweni yomtholampilo.Isibonelo, ucwaningo olulodwa lweshabhu lwahlaziya uhlelo 13 lwe-Google olususelwa ku-AI lokutholwa kwesifo sikashukela ukuze kuhlonzwe izinselele endleleni esuka elebhu iye emtholampilo, njengezidingo zokuqinisekisa zangaphandle kanye nezindlela zokugunyaza ezilawulayo.
Sebenzisa ukufunda kokuhlangenwe nakho: Amakhono obuchwepheshe adinga ukuzijwayeza okugxilile nokusebenzisa okuphindaphindiwe ukuze abe yingcweti, okufana nolwazi lokufunda oluzungezisayo lwabaqeqeshwayo basemtholampilo.Isixazululo esisodwa esingaba khona imodeli yekilasi ephenyisiwe, okuye kwabikwa ukuthi ithuthukisa ukugcinwa kolwazi emfundweni yobunjiniyela14.Kulo modeli, abafundi babuyekeza izinto zetiyori ngokuzimela futhi isikhathi sekilasi sinikelwa ekuxazululeni izinkinga ngezifundo eziyisibonelo.
Ukukala kwabahlanganyeli bemikhakha eminingi: Sibona ngeso lengqondo ukwamukelwa kwe-AI okubandakanya ukubambisana kuzo zonke iziyalo eziningi, okuhlanganisa odokotela kanye nochwepheshe bezempilo abahlangene abanamazinga ahlukahlukene okuqeqeshwa.Ngakho-ke, amakharikhulamu angase adinge ukuthuthukiswa ngokubonisana nobuhlakani beminyango ehlukene ukuze okuqukethwe kwawo kuhambisane nezindawo ezahlukene zokunakekelwa kwezempilo.
I-Artificial intelligence iyi-high-tech futhi imiqondo yayo eyinhloko ihlobene nezibalo nesayensi yekhompyutha.Ukuqeqesha abasebenzi bezokunakekelwa kwempilo ukuze baqonde ubuhlakani bokwenziwa kuletha izinselele eziyingqayizivele ekukhetheni okuqukethwe, ukuhambisana komtholampilo, nezindlela zokulethwa.Sithemba ukuthi ukuqonda okutholwe kuma-workshops e-AI kwezemfundo kuzosiza othisha bakusasa bamukele izindlela ezintsha zokuhlanganisa i-AI emfundweni yezokwelapha.
Iskripthi se-Google Colaboratory Python siwumthombo ovulekile futhi sitholakala ku-: https://github.com/ubcaimed/ubcaimed.github.io/tree/master/.
U-Prober, u-KG no-Khan, S. Ukucabanga kabusha ngemfundo yezokwelapha: ukubizelwa esenzweni.Akkad.umuthi.88, 1407–1410 (2013).
McCoy, LG njll. Yini ngempela abafundi bezokwelapha okudingeka bayazi mayelana nobuhlakani bokwenziwa?Izinombolo ze-NPZh.Umuthi 3, 1–3 (2020).
U-Dos Santos, DP, et al.Izimo zengqondo zabafundi bezokwelapha ngobuhlakani bokwenziwa: inhlolovo yezikhungo eziningi.I-EURO.imisebe.29, 1640–1646 (2019).
Umlandeli, u-KY, u-Hu, u-R., no-Singla, R. Isingeniso sokufunda ngomshini kubafundi bezokwelapha: iphrojekthi yokuhlola.J. Med.fundisa.54, 1042–1043 (2020).
Cooperman N, et al.Ukuhlonza izingane ezisengozini ephansi kakhulu yokulimala kobuchopho okubalulekile emtholampilo ngemva kokulimala ekhanda: ucwaningo lweqembu oluzoba yiqembu.I-Lancet 374, 1160–1170 (2009).
Street, WN, Wolberg, WH kanye neMangasarian, OL.Ukukhishwa kwesici se-nuclear sokuxilongwa kwesimila sebele.Isayensi Yezinto eziphilayo.Ukucutshungulwa kwesithombe.Isayensi Yezinto eziphilayo.Weiss.1905, 861–870 (1993).
I-Chen, i-PHC, i-Liu, i-Y. kanye ne-Peng, L. Ungawakha kanjani amamodeli wokufunda ngomshini wokunakekelwa kwezempilo.Nat.Math.18, 410–414 (2019).
Selvaraju, RR et al.I-Grad-cam: Ukuhunyushwa okubonakalayo kwamanethiwekhi ajulile kusetshenziswa ukwenziwa kwasendaweni okususelwa ku-gradient.Izinqubo zeNgqungquthela Yamazwe Ngamazwe ye-IEEE yeComputer Vision, 618–626 (2017).
U-Kumaravel B, uStewart K kanye no-Ilic D. Ukuthuthukiswa nokuhlolwa kwemodeli ezungezayo yokuhlola amakhono emithi asekelwe ebufakazini kusetshenziswa i-OSCE emfundweni yezokwelashwa yeziqu ze-undergraduate.Imithi ye-BMK.fundisa.21, 1–9 (2021).
I-Kolachalama VB kanye ne-Garg PS Machine yokufunda kanye nemfundo yezokwelapha.Izinombolo ze-NPZh.umuthi.1, 1–3 (2018).
van Leeuwen, KG, Schalekamp, ​​​​S., Rutten, MJ, van Ginneken, B. kanye no-de Rooy, M. Artificial intelligence ku-radiology: Imikhiqizo yezohwebo eyi-100 kanye nobufakazi bayo besayensi.I-EURO.imisebe.31, 3797–3804 (2021).
Topol, EJ Umuthi osebenza kahle kakhulu: ukuhlangana kobuhlakani bomuntu nobokwenziwa.Nat.umuthi.25, 44–56 (2019).
Bede, E. et al.Ukuhlola okugxile kumuntu kwesistimu yokufunda ejulile esetshenziswe emtholampilo ukuze kutholwe i-retinopathy yesifo sikashukela.Izinqubo zeNgqungquthela ye-CHI yango-2020 Yezici Zabantu ku-Computing Systems (2020).
Kerr, B. Ikilasi eliphendukile emfundweni yobunjiniyela: Ukubuyekezwa kocwaningo.Izinqubo zeNgqungquthela Yamazwe Ngamazwe ka-2015 Yokufunda Ngokuhlanganyela Ngokuhlanganyela (2015).
Ababhali babonga uDanielle Walker, u-Tim Salcudin, no-Peter Zandstra abavela ku-Biomedical Imaging and Artificial Intelligence Research Cluster e-University of British Columbia ngokusekela nokuxhaswa ngemali.
I-RH, PP, ZH, RS kanye ne-MA babenomthwalo wemfanelo wokuthuthukisa okuqukethwe kokufundisa kweshabhu.I-RH ne-PP babenomthwalo wemfanelo wokuthuthukisa izibonelo zezinhlelo.I-KYF, i-OY, i-MT kanye ne-PW babenomthwalo wemfanelo wokuhleleka kwephrojekthi kanye nokuhlaziywa kwama-workshops.I-RH, OY, MT, RS yayinesibopho sokudala izibalo namathebula.U-RH, KYF, PP, ZH, OY, MY, PW, TL, MA, RS babenomthwalo wemfanelo wokubhala nokuhlela idokhumenti.
I-Communication Medicine ibonga uCarolyn McGregor, Fabio Moraes, no-Aditya Borakati ngeqhaza labo ekubuyekezweni kwalo msebenzi.


Isikhathi sokuthumela: Feb-19-2024