JAGORAN AL'UMMA
Bias ta atomatik a cikin Clinical AI
Son zuciya ta atomatik shine hali don fifita shawarar kwamfuta akan hukuncin ku ko akan sabanin hujja.
A wannan shafi4 min karatu
Dubawa
A cikin AI na asibiti yana jagorantar likitocin su bi abubuwan da ba daidai ba (kurakurai na hukumar) ko rasa matsalolin tsarin bai yi alama ba (kurakurai na tsallakewa). Yana da mahimmanci saboda amincin kayan aikin AI na ainihin-duniya ya dogara da yadda mutane ke amfani da shi, kuma duka fiye da amincewa da rashin amincewa na iya soke fa'idar ingantaccen samfuri.
Zurfafa nutsewa
An fara nazarin son zuciya ta atomatik sosai a cikin jirgin sama da sauran ayyuka masu mahimmancin aminci. Masu binciken abubuwan ɗan adam sun gano cewa ma'aikatan da ke aiki tare da ingantattun ingantattun kayan aiki suna sa ido kan ƙasa da jinkirtawa. Sun bayyana nau'ikan kuskure guda biyu. A cikin kuskuren tsallakewa, mutum ya rasa matsala saboda tsarin bai nuna shi ba. A cikin kuskuren kwamiti, mutum ya yi aiki da wata shawara mara kyau. Wani bita na tsari na 2012 da Goddard da abokan aiki suka samu ta atomatik a cikin goyan bayan yanke shawara na asibiti, tare da tasirin da aka tsara ta hanyar rikitarwar ɗawainiya, aikin aiki, ƙwarewa da amincewa. Hoto na likita yana ba da mafi kyawun shaida. Binciken da aka yi na gano farkon da kwamfuta ta taimaka a mammography ya ruwaito cewa lokacin da tsarin ya kasa haifar da ciwon daji, wasu lokuta masu karatu ba su iya samun ta fiye da yadda za su kasance ba tare da tsarin ba. Wani bincike na 2023 a cikin Radiology ya bai wa masana rediyo da gangan shawarwarin da ba daidai ba da aka gabatar kamar yadda suke fitowa daga AI lokacin tantance mammograms. Daidaiton su ya ragu, kuma ƙwararrun masu karatu sun fi shafa. Wani bincike na vignette na 2023 a cikin JAMA ya gano cewa likitocin sun nuna tsarin nuna son kai na tsari wanda ya yi rashin ingantaccen bincike, kuma ƙara bayanin samfurin bai yi kadan don magance cutar ba. Akwai kuma akasin matsalar. Bayan ganin algorithm yana yin kuskure, mutane na iya daina amfani da shi ko da ya fi su. Masu bincike a wajen likitanci sun kira wannan ƙiyayyar algorithm. A asibitoci yakan yi kama da gajiyawar faɗakarwa. Duk matsananci ba shi da lafiya. Manufar ita ce dogaro mai dacewa: amincewa da kayan aiki a inda yake da karfi da kuma duba inda yake da rauni. Mahimman fahimta guda biyu: sanya "mutum cikin madauki" yana tabbatar da aminci. Idan ɗan adam ya dogara da jinkiri, madauki yana ƙara ƙaramin kariya; kuma Bayani koyaushe yana taimakawa. Wasu bincike sun nuna suna iya ƙara dogaro ga abubuwan da ba daidai ba cikin sauƙi kamar waɗanda suke daidai.
Dabarun Tasiri
Haɗari da aminci
Bala'i da cutar AI ta yau da kullun duka sun dogara da wanda ya fahimci haɗarin kuma wanda zai iya yin aiki.
Shawarwari masu haske
Ilimin jama'a da na ƙwararru yana siffanta ko ƙaƙƙarfan manufofin aminci na yiwuwa a siyasance.
Yanke ta hanyar yayatawa
Bayyanar bayani yana rage kama ta hanyar zage-zage, dakin gwaje-gwaje PR, da gidan wasan kwaikwayo mara kyau.
Makomar Bias Automation a Clinical AI
Kamar yadda kayan aikin AI ke bazuwa a cikin hoto, takaddun bayanai da rarrabewa, son zuciya ta atomatik yana zama matsala ta ƙira da gudanarwa gami da mutum ɗaya. Masu bincike suna nazarin waɗanne zaɓin mu'amala, hanyoyin horo da madaukai na amsa suna tallafawa dogaro da ya dace. Yawancin sakamako ya zuwa yanzu sun fito ne daga nazarin kwaikwayi ko vignette, don haka shaidar duniya ta kasance mai iyaka. Tsarin kiwon lafiya sun fara bin tsarin ƙetare da karɓa azaman alamun aminci. Wataƙila ƙalubalen da ke gaba shine kare ƙwarewar masu zaman kansu tsakanin masu horarwa waɗanda ke koyo da AI tun daga farko, don haka har yanzu suna iya kama kurakurai lokacin da kayan aikin ya gaza.
Aiwatar da Gaskiyar Duniya
Likitan rediyo da ke karanta mammogram bai ga alamar AI ba a kan wata dabarar asymmetry kuma ta ci gaba, kodayake da ta yi alama da kanta. Wannan kuskure ne na tsallakewa.
Likitan harhada magunguna ya ajiye damuwarsa game da kashi na koda saboda tsarin tabbatar da oda ya tayar da hankali. Bai gane cewa sabon sakamakon creatinine na mara lafiya bai taɓa kaiwa ga tsarin ba.
Bayan makonni na ƙararrawa na ƙarya, likitan gaggawa ya kori kowane faɗakarwar sepsis kuma ya yi watsi da wanda yake daidai. Wannan gajiya ce mai faɗakarwa da rashin amana.
Mazauni mai ilimin fata yana canza madaidaicin ganewar asali na melanoma zuwa nevus mara kyau bayan wani app ya yiwa rauni rauni. Rahoton Pathology daga baya ya saba wa ka'idar. Wannan kuskure ne na hukumar.
Hatsari & Tsare-tsare
Magance haɗarin wanzuwa azaman sci-fi yayin da abubuwan iyawa.
Amintaccen samfur mai ruɗani tare da jeri ƙarƙashin babban ikon kai.
Barin waɗanda ba Ingilishi ba da ƙwararrun masu sauraro tare da tushe masu ƙarancin inganci kawai.
Taswirar Hanya
Rarrabe lahani na samfur, rashin amfani, da hasarar sarrafa-haɗari / rashin daidaituwa.
Tambayi wane shaida zai canza ra'ayin ku akan jerin lokuta da tsanani.
Fi son tushe na farko da tabbataccen kimantawa akan da'awar tallace-tallace.
Gano hanyar aiki ɗaya: aiki, manufa, kuɗi, ko ƙwarewa - ba kawai sani ba.
Ci gaba da Bincike
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Automation Bias in Clinical AI quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Tambayoyin da ake yawan yi
Menene Bias Automation a Clinical AI?
Son zuciya ta atomatik shine hali don fifita shawarar kwamfuta akan hukuncin ku ko akan sabanin hujja. A cikin AI na asibiti yana jagorantar likitocin su bi abubuwan da ba daidai ba (kurakurai na hukumar) ko rasa matsalolin tsarin bai yi alama ba (kurakurai na tsallakewa). Yana da mahimmanci saboda amincin kayan aikin AI na ainihin-duniya ya dogara da yadda mutane ke amfani da shi, kuma duka fiye da amincewa da rashin amincewa na iya soke fa'idar ingantaccen samfuri.
Likitan rediyo ya rasa dabarar asymmetry saboda AI ba ta sanya alama a wurin ba, kodayake da ta kama shi da kanta. Wane irin kuskuren son zuciya ne wannan?
Kuskuren tsallakewa ya rasa matsala saboda tsarin bai yi alama ba. Kuskuren hukumar yana aiki akan ba daidai ba shawara.
Menene binciken Radiology na 2023 ya gano lokacin da masana rediyo suka karɓi shawarwarin da aka yiwa lakabi da AI da gangan akan mammograms?
Shawarwarin da ba daidai ba sun rage daidaito, kuma ƙwararrun likitocin rediyo sun fi saurin kamuwa da su. Wannan yana nuna cewa gwaninta yana taimakawa amma ba ya cika karewa.
Bisa ga taƙaitaccen jagorar na nazarin vignette na JAMA na 2023, menene ya faru lokacin da aka ƙara bayani zuwa tsari na son rai?
Binciken ya gano cewa samfurori masu banƙyama sun rage daidaito kuma bayanin ya ba da kariya kaɗan. Wannan yana ƙalubalanci imani cewa bayani koyaushe yana taimakawa.
Bayan makonni na faɗakarwar sepsis na ƙarya, likita ya kori duka, gami da daidai. Menene wannan ya kwatanta?
Maimaita ƙararrawa na ƙarya na iya tura likitocin asibiti suyi watsi da kayan aiki gaba ɗaya, wanda shine akasin gazawar daga yawan amana. Masu bincike a wajen likitanci suna kiran ƙiyayya ta algorithm mai alaƙa.
Wanne ƙirar aikin aiki ne jagorar ya ce mafi kyawun kiyaye hukunci mai zaman kansa?
Lokacin da likitocin suka fara yanke hukunci, AI yana aiki azaman ra'ayi na biyu maimakon anka wanda ke siffanta ra'ayinsu na farko.
Ci gaba da koyo
Jagora masu alaƙa
An zaɓi ƙarin jagora don wannan batu