HejuruUbuyobozi bukurikira
AI Kugwa Ingaruka Zishobora Kugaragara Mubitaro
Inganda
UMUYOBOZI W'umuryango
Guteganya umwanya w'ubuzima bwo kwica ubwawe hakoreshejwe AI bikoresha imibare n'imodere z'ubuhanga bwa mashini zizizwe ku makuru y'ubuzima kugira ngo zitegereze abarwayi bafite amahirwe menshi yo gukora igikorwa cyo kwica ubwawe cyangwa gupfa mu minsi iri imbere, kugira ngo sisitemu z'ubuzima zishobore gutanga ubutabazi n'ubushakashatsi ku bafite umwanya mwinshi.
Umva iki gitaboAI-byavuzwe · ~3 min
Ibi birakenewe kuko abantu benshi bapfa bakiyica babonye umuganga mu mwaka wabanje, kandi ibitekerezo by'abaganga batabikoresheje byagaragaye ko bifite amahirwe make gusa kurusha ibisanzwe. N'imodere nziza cyane zigaragaza abantu benshi batagomba kwica ubwawe kurusha ababigomba.
Most health system models draw on data already in the electronic health record: prior suicide attempts and self-harm diagnoses, mental health and substance use diagnoses, prescriptions, emergency visits and inpatient stays, and screening answers such as item 9 of the PHQ-9, which asks about thoughts of death or self-harm. Methods range from penalized logistic regression to random forests and gradient-boosted trees. Well-known efforts include models from the Mental Health Research Network, which includes Kaiser Permanente sites, work at Vanderbilt University Medical Center, and the VA's REACH VET, launched in 2017, which flags roughly the top 0.1 percent of risk scores for review. Several large studies report good discrimination, meaning the models rank higher-risk patients above lower-risk ones reasonably well. The main misconception is that these models predict who will die. They do not. Suicide is rare in any given month, so even an accurate model produces mostly false positives; most flagged patients will never attempt. The models are best understood as tools for prioritizing limited clinical attention. That makes the response to a flag the real ethical question. Good practice treats a flag as a reason for caring, voluntary outreach: a check-in, a review of the care plan, safety planning, and discussing limits on access to lethal means such as safe firearm storage. Poor practice uses flags punitively or coercively, shares them outside clinical care, or leaves them unexplained to patients. Health systems also need to test performance across racial, ethnic, age and sex groups, because a model trained on past records inherits gaps in who was diagnosed and treated. Outside health care, platforms such as Facebook have used automated detection of suicidal posts since 2017, with much less public validation.
Porogaramu ya REACH VET ya Leta ya Amerika y'Abasirikare bageze mu kigero (VA) itanga amanota ku basirikare bageze mu kigero bari mu kurekurwa kwa VA buri kwezi, kandi umuyobozi w'ahantu asuzuma abafite umwanya mwinshi hamwe n'abaganga babo kugira ngo barebe imishinga y'ubuvuzi bagahindura uburyo bwo kubafata.
Sisitemu y'ubuzima igaragaza amanota y'umwanya mu biganiro by'ubuzima bw'umutima by'abarwayi bataba mu bitaro, bigatuma umuganga akora ubushakashatsi bwihariye nka Columbia Suicide Severity Rating Scale aho kwizerwa amanota gusa.
Ibitaro by'ibikorwa byihutirwa bikoresha ikimenyetso cy'imodere kugira ngo bategure igikorwa cy'umutekano mbere y'uko umwarwayi ava, bikurikirwa n'ubufatanye bw'umutima nka ubutumwa bw'ibanze bwo gusubiramo.
Sisitemu y'ubuzima isuzuma imodere yayo, isanga ikora nabi ku itsinda rimwe ry'abantu, ikayisubiramo mbere yo kwagura uburyo bwo kuyikoresha.
Abashakashatsi barimo gushakisha ibindi bikoresho byinshi, nka ibyanditswe by'ubuvuzi bikorwa n'imodere z'ururimi, hamwe n'amakuru asanzwe, nubwo ibikoresho byinshi byongera ibibazo by'umutekano w'ibanga kandi ntibigire icyizere cy'ibyavuyemo byiza. Ikibazo cy'ingenzi gisigaye si ukwemeza imodere ahubwo ni ingaruka: niba porogaramu zishingiye ku bimenyetso zigabanya igikorwa cyo kwica ubwawe n'urupfu ugereranyije n'ubuvuzi busanzwe bwiza. Tekereza ko hazaba umwitozo mwinshi ku isuzuma ry'uburinganire, ubusobanuro n'abarwayi no guhuza imodere n'ibikorwa byemejwe nka gahunda y'umutekano n'ubufatanye.
Ibaruwa y'amakuru y'ubuntu
Inkuru eshatu zizewe za AI buri gitondo cy'iminsi y'akazi, zanditse mu rurimi rworoshye. Ubuntu iteka ryose, nta kwamamaza.
Imeri imwe buri munsi y'akazi. Kwiyimura mu buryo bumwe. Ntitwigera tugurisha cyangwa ngo dusangire aderesi yawe.
Isuzume
Ibisubizo ako kanya kuri buri kibazo, n'icyemezo gishobora gusangirwa gifite ID yagenzurwa umaze gutsinda isomo.
Fasha uburezi bwa AI bw'ubuntu. AI Understanding ni umuryango utegamiye kuri leta wa 501(c)(3) — nta kwamamaza, nta kwishyura, iteka ryose. Tanga impano
Guteganya umwanya w'ubuzima bwo kwica ubwawe hakoreshejwe AI bikoresha imibare n'imodere z'ubuhanga bwa mashini zizizwe ku makuru y'ubuzima kugira ngo zitegereze abarwayi bafite amahirwe menshi yo gukora igikorwa cyo kwica ubwawe cyangwa gupfa mu minsi iri imbere, kugira ngo sisitemu z'ubuzima zishobore gutanga ubutabazi n'ubushakashatsi ku bafite umwanya mwinshi. Ibi birakenewe kuko abantu benshi bapfa bakiyica babonye umuganga mu mwaka wabanje, kandi ibitekerezo by'abaganga batabikoresheje byagaragaye ko bifite amahirwe make gusa kurusha ibisanzwe. N'imodere nziza cyane zigaragaza abantu benshi batagomba kwica ubwawe kurusha ababigomba.
Iyibona ibintu 90 nyazo ariko nanone ikagaragaza abantu 9,990 batagomba kwica ubwawe, bityo agaciro k'ibimenyetso byiza kagabanuka munsi ya 1%.
Kubera ko kwica ubwawe ari ibintu bidasanzwe, n'imodere zemeza zigaragaza abantu benshi batagomba kwica ubwawe. Zifasha mu gufata icyemezo cy'aho gutanga ubutabazi, aho kumenya uwo azapfa.
REACH VET, yatangiye mu 2017, yagaragaza hafi 0.1% y'amanota yo hejuru kugira ngo isuzumwe n'umuyobozi w'ahantu n'abaganga.
Ikimenyetso kiba gifite akamaro gusa niba umuntu ashobora gukora icyo kibazo, bityo amakuru y'umwanya ahujwe n'abakozi b'ubutabazi bariho.
Uburyo bwiza bufata ikimenyetso nk'impamvu yo gufatanya n'umutima, harimo no gukingira ibikoresho bishobora gukoreshwa mu kwica ubwawe, aho gukora ibikorwa by'ihohoterwa cyangwa gushimuta.
Komeza wige
Abandi bayobozi batoranijwe kuriyi ngingo
HejuruUbuyobozi bukurikira
AI Kugwa Ingaruka Zishobora Kugaragara Mubitaro
Inganda