Isifinyezo: Ucwaningo lokuqala, lokuqala lwalo lwabonisa ukuthi uhlelo oluqondene nomuntu siqu, oluqondiswa umshini lokufunda indlela yokuphila lungacishe luphinde kabili izilinganiso zokuxolelwa kokudangala okumaphakathi kuya kokuphakathi. Ucwaningo lulandelela ukuthi izici zokuziphatha ngazinye zibikezela ngokuhlukile izimo zemizwa ephansi.
Ngokusebenzisa ama-smartwatches omthengi kanye namalogi edatha esikhathi sangempela ukuze bakhe uhlelo lomuntu ngamunye lokwandisa imizwa (i-iMAP), abaphenyi bathole izinga lokuxolelwa kokudangala elingu-55% kanye nokwehliswa okuphawulekayo kokukhathazeka, okuhlinzeka ngohlaka olusebenza kahle kakhulu lokunakekelwa kwengqondo komuntu siqu, okukude.
Amaqiniso Abalulekile
- Ukwehluleka Ukuphindaphinda: Bangaphezu kuka-21% abantu abadala base-US baphila nokucindezeleka. Nakuba imihlahlandlela evamile yomtholampilo yeluleka ukulungiswa okuvamile kokulala, ukujima, nokudla, lezi zincomo ezilingana nazo zonke zivame ukukhungathekisa abantu abacindezelekile futhi zehluleke ngenxa yokuhlukahluka okuphezulu komuntu nomuntu.
- Ukuhlolwa Kwebhayomethrikhi Yamaviki Amabili: Ukuze kwakhiwe isisekelo esiqhutshwa yidatha, abantu abadala abangu-50 abanokucindezeleka okumaphakathi kuya kokulingene bagqoke amawashi ahlakaniphile ukuze balandelele ukushaya kwenhliziyo yabo nokunyakaza komzimba kuyilapho beloga amamethrikhi asendaweni asendaweni ngekhwalithi yokulala, ukudla, nokusebenzelana komphakathi kufika izikhathi ezine ngosuku.
- Isu le-iMAP: I-UC San Diego's Neural Engineering and Translation Labs (NEATLabs) isebenzise le datha yomuntu siqu ngemodeli yokufunda yomshini ukuze ihlukanise izingcupho eziphezulu zendlela yokuphila ezishayela isimo sokudangala sombambi qhaza ngamunye. Abaqeqeshi bezempilo babe sebehlanganisa le mininingwane nemithi yokwelapha ehambisanayo, esekelwe ebufakazini ukuze bakhe uhlelo lokukhuliswa kwemizwa yomuntu ngamunye (i-iMAP).
- Ukuphindwa kabili kwezimpawu zomtholampilo: Ukungenelela okujwayelekile kokuziphatha okumaphakathi kwezinga lokuxolelwa elingama-30%. Ngokuphambene, indlela ye-algorithmic iMAP iqalise a 55% izinga lokuxolelwaokusho ukuthi ngaphezu kwengxenye yeqembu alisakwazi ukuhlangabezana nemibandela yomtholampilo yokucindezeleka ngemva kwamasonto ayisithupha.
- Izabelo ze-Peripheral Health: Ngaphandle kokwelapha izimpawu zokucindezeleka zendawo, uhlelo lokufundisa ngevidiyo olukude lwamasonto ayisithupha lubangele Ukwehla okungu-36% kwezimpawu zokukhathazekaukuthuthukiswa okukhulu kwekhwalithi yempilo ezibike yona, kanye nezikolo eziphakeme ekuhloleni inkumbulo emfushane nokunaka.
- Umthelela Oqhubekayo Wokwelapha: Ukucwaninga kokulandelela kuqinisekisile ukuthi izinzuzo zengqondo nezengqondo ezitholwe ngesikhathi sokuqeqeshwa kwamasonto ayisithupha ziphikelele izinyanga ezintathu ezigcwele ngemva kokuphela kokungenelela okusebenzayo.
Umthombo: UCSD
Bangaphezu kuka-21% abantu abadala base-US abacindezelekile, okuba nomthelela omkhulu kukhwalithi yabo yempilo. Abantu abaningi abanokucindezeleka okumaphakathi kuya kokuphakathi bangakwazi ukuthuthukisa izimpawu zabo ngokulungisa imikhuba yansuku zonke efana nokulala, ukuvivinya umzimba, ukudla nokuxhumana nabantu, ngokusho kuka-Jyoti Mishra, PhD, uprofesa ohlobene wezengqondo e-University of California San Diego School of Medicine.
Kodwa-ke, ngenxa yokuthi ukudangala kuyahlukahluka kakhulu phakathi kwabantu, indlela yokuphila elingana nosayizi owodwa ayisebenzi kahle.
Ocwaningweni lokuqala lwayo lohlobo, uMishra nethimba lakhe bathuthukise uhlelo lokuqeqesha oluqondiswa umshini olusekelwe kudatha eqoqwe ngamadivaysi omuntu siqu mayelana nesimo sengqondo sabahlanganyeli kanye nemikhuba yansuku zonke.
Bathole ukuthi ababambiqhaza abasebenzisa lolu hlelo bathola ukuncipha okukhulu kwezimpawu zokucindezeleka ngemva kwamasonto ayisithupha. Okutholakele kunikeza indlela ethembisayo yokuletha ukude ukwelashwa komuntu siqu kokucindezeleka okuklanyelwe izimo zomuntu ngamunye.
Ucwaningo lushicilelwe ku I-NPP – I-Digital Psychiatry kanye ne-Neuroscience.
Ngesikhathi samaviki amabili, abantu abadala abangu-50 abanokucindezeleka okumaphakathi kuya kokumaphakathi babegqoke iwashi elihlakaniphile elilandelela izinga lokushaya kwenhliziyo namazinga okuzivocavoca. Ngaphezu kwalokho, bafaka imizwa yabo futhi baphendula imibuzo emifushane kuze kube kane ngosuku mayelana nekhwalithi yabo yokulala, ukudla, izinga lomsebenzi, nokuthi bakhuluma kangaki nabangane noma umndeni.
Ithimba lenze imodeli yokufunda yomshini ehlukile kumhlanganyeli ngamunye ngokususelwe kule datha ukuze lithole ukuthi yiziphi izici zendlela yokuphila ezibikezele kangcono isimo esiphansi somuntu. Bese, umhlanganyeli ngamunye wasebenza nomqeqeshi wezempilo ukuze asebenzise uhlelo lomuntu ngamunye lokukhuliswa kwemizwa, noma i-iMAP.
“Inhloso yethu bekuwukuthola izinto eziphambili zokuphila eziqhuba isimo sokucindezeleka, ezingahluka kubantu abahlukene, kanye nokuthola ukuthi ngokukhomba lezo zici ngokuqeqeshwa komuntu siqu, abantu bazozizwa bengcono ngempela,” kusho uMishra, oqondisa ngokubambisana kwe-UC San Diego's Neural Engineering and Translation Labs (NEATLabs).
Emavikini ayisithupha alandelayo, ababambiqhaza basebenze nomqeqeshi wabo ukusebenzisa ama-iMAPs abo.
“Wonke umuntu ovivinyweni wayesebenzisa izindlela zokwelapha ezehlukene esezivele zibhalwe kahle ezincwadini ngokuya ngesici esiqagelayo,” kusho uMishra. “Abanye abantu bebesebenza ngohlelo lokwelapha ngokuziphatha kwengqondo ngenxa yokuqwasha, abanye bekhulisa izinhlobo zemisebenzi yokuzivocavoca abebevele beyenza empilweni yabo yansuku zonke, ukuthuthukisa ukuxhumana nomphakathi, noma ukusebenzisa ukungenelela okusekelwe ekudleni okunempilo.”
Ngemuva kokusebenza nomqeqeshi ngamakholi wevidiyo amafushane amasonto ayisithupha, ababambiqhaza:
- Kubikwe ukwehliswa okukhulu kwezimpawu zokucindezeleka; U-55% wabahlanganyeli akasenawo ukucindezeleka ngemva kokwelashwa njengoba kukalwa yi-Patient Health Questionnaire-9 (PHQ-9), ukuhlolwa kokuhlolwa kokucindezeleka okujwayelekile.
- Kubikwe ukwehla okungama-36% kwezimpawu zokukhathazeka njengoba kulinganiswa ukuhlolwa kokuhlolwa kwe-Generalized Anxiety Disorder-7 (GAD-7).
- Ubike ukuthuthukiswa okuphawulekayo kukhwalithi yabo yempilo.
- Ithole amaphuzu aphezulu ekuhlolweni kwenkumbulo emfushane nokunaka.
Ngaphezu kwalokho, abacwaningi bathola ukuthi umphumela wokwelapha waqhubeka phakathi nezinyanga ezintathu abaqhubeka belandela abahlanganyeli ngemva kokuphela kokungenelela.
“Izivivinyo zomtholampilo zibonisa ukuthi ukungenelela okuningi kwamanje kukhombisa kuphela inzuzo engama-30% ngokwesilinganiso ngokuya ngokuxolelwa kokucindezeleka; lapha sibona ukuphindaphinda okucishe kube kabili kwalokho ngenxa yokukhomba izinto eziqagela indlela yokuphila ephezulu ngokuqeqeshwa komuntu siqu okuqhutshwa yidatha,” kusho uMishra.
UMishra ucabanga ukuthi ukungenelela kungase kusebenze kakhulu ngoba kuwukusuka ezincomweni ezijwayelekile zempilo yokuziphatha.
“Wonke umuntu uyazi ukuthi kufanele sidle ukudla okunempilo noma sizame ukulala amahora ayisishiyagalombili noma sizivocavoce imizuzu engu-150 ngesonto nokunye,” kusho yena. “Kodwa ngicabanga ukuthi imininingwane yomuntu siqu ingaba namandla kakhulu kunalezi ziqondiso ezijwayelekile ngoba azikhuzi kangako. Uma umuntu esesimweni sokucindezeleka, akwenzeki ukuguqula yonke into ngempilo yakhe – uzama nje ukusinda futhi usebenze usuku nosuku.”
Nakuba lolu cwaningo beluluncane, lunikeza ubufakazi bokuqala bokuthi ukuqapha kwedijithali, imininingwane etholakala emshinini kanye nokuqeqeshwa okufushane, komuntu siqu kwamasonto onke okulethwa kude kungase kube indlela ehlanganisiwe ethembisayo yokubhekana nokucindezeleka okumaphakathi kuya kokuphakathi emaqenjini amakhulu abantu. Ucwaningo olukhudlwana, olulawulwayo lwale ndlela yokwelapha yomuntu siqu iyadingeka ukuze kuqinisekiswe okutholakele.
Ababhali ababambisene abengeziwe ocwaningweni bahlanganisa: Jason Nan, Suzanna Purpura, Satish Jaiswal, Houtan Afshar, Vojislav Maric, James K. Manchanda noCharles T. Taylor e-UC San Diego; kanye no-Dhakshin Ramanathan e-UC San Diego nase-VA San Diego Medical Center.
Uxhaso: Ucwaningo luxhaswe ingxenye yemali yembewu evela kwaHope for Depression Research Foundation.
Imibuzo Ebalulekile Iyaphendulwa:
A: Uma umuntu evaleleke esimweni sokucindezeleka, usebenza kumodi yokusinda ezama nje ukusebenza usuku nosuku. Ukubanika uhlu olukhulu, olujwayelekile lwezinguquko zendlela yokuphila kungaphezu kwamandla. Ngenxa yokuthi ukucindezeleka kuyahlukahluka phakathi kwezinto eziphilayo, lokho okusebenzela umuntu oyedwa ngeke kumsebenzele omunye. Iqondise imbangela eyodwa, esekelwe idatha lakho imizwa ethile ephansi isusa ukukhathala kokuzama ukushintsha yonke into ngesikhathi esisodwa.
A: Amaviki amabili, imodeli ilandelela amaphethini phakathi kokunyakaza kwakho kwebhayomethrikhi kanye namalogi akho emizwa asebenzayo. Ibala ukuhlobana okuyisisekelo ukuze kutholakale isici sakho esiyinhloko sobungozi. Uma i-algorithm ithola ukuthi ukuntuleka kokuxhumana nabantu yikho okubikezela ngokuqondile isimo sakho sengqondo esiphansi, umqeqeshi wakho ukunika izindlela zokwelapha zokuthuthukisa ukuxhumana nabantu. Uma ukulala kwakho kungumshayeli oyinhloko, ufaka ithrekhi yokwelashwa kokuziphatha okuqondisiwe yokuqwasha.
A: Idatha ibonisa ukuphulukiswa kwakhelwe ukuhlala. Ukuhlolwa komtholampilo kwembule ukuthi ukwehla okukhulu kokucindezeleka nokukhathazeka, kanye nokuthuthukiswa kwenkumbulo nokunaka, kwahlala kuzinzile ngokuphelele phakathi nenkathi yokulandelela yezinyanga ezintathu ngemva kokuphela kwezingcingo zokuqeqesha ividiyo zamasonto onke.
Amanothi Omhleli:
- Lesi sihloko sihlelwe umhleli weNeuroscience News.
- Iphepha lejenali libuyekezwe ngokugcwele.
- Ingqikithi eyengeziwe yengezwe abasebenzi bethu.
Mayelana nale AI kanye nezindaba zocwaningo lokucindezeleka
Umbhali: Susanne Bard
Umthombo: UCSD
Othintana naye: USusanne Bard – UCSD
Isithombe: Isithombe sifakwe ku-Neuroscience News
Ucwaningo lwangempela: Vula ukufinyelela.
“Ukungenelela okuqondisiwe komshini wokufunda ngomshini wokuthuthukisa indlela yokuphila ekucindezelekeni: isifundo somshayeli” ka-Jason Nan, Suzanna Purpura, Satish Jaiswal, Houtan Afshar, Vojislav Maric, James K. Manchanda, Charles T. Taylor, Dhakshin Ramanathan & Jyoti Mishra. I-NPP—I-Digital Psychiatry ne-Neuroscience
I-DOI:10.1038/s44277-026-00062-3
Abstract
Ukufunda komshini komuntu siqu ukungenelela okuqondisiwe kokuthuthukisa impatho yendlela yokuphila ekucindezelekeni: isifundo sokulinga
Ukungenelela komuntu siqu okuqhutshwa yidatha yokucindezeleka kuyadingeka kakhulu. Lapha, sisebenzise ukufundwa komshini we-N-of-1 (ML) ukuze siqondise ngokugcwele ukungenelela kwendlela yokuphila yokucindezeleka. Abantu abangu-50 abanokucindezeleka okumaphakathi kuya kokuphakathi ababhalise ingalo eyodwa, ilebula evulekile ye-Personalized Mood Augmentation (PerMA) pilot clinical trial (NCT05662254).
Ababambiqhaza baphothule isigaba sokuqapha kwedijithali samasonto amabili besebenzisa ukuhlola kwesikhashana kwemvelo okusekelwe ku-smartphone (ama-EMA, 4×/ngosuku) kanye nokulandelela iwashi elihlakaniphile lokulandela imizwa nezici zokuphila (ukulala/ukuvocavoca/ukudla/ukuxhumana nomphakathi).
Amamodeli e-ML aqondene nawe akhiqizwa kusukela kule datha ukuze kuhlonzwe izici zendlela yokuphila ezibikezela kakhulu isimo somuntu ngamunye, futhi imiphumela yahunyushelwa ezinhlelweni zokukhulisa isimo somuntu ngamunye (ama-iMAP) ezisetshenziswa ababambiqhaza amaviki ayisithupha ngokuqondisa komqeqeshi wezempilo kanye ngesonto.
Abaqedi bokungenelela (n = 40) babonise ukwehla okuphawulekayo kwezimpawu zokucindezeleka (umphumela oyinhloko ozilinganisela ngokwakho nge-PHQ9 −3.5 ± 3.8, Cohen's d = −0.89, CI [−1.25 −0.53]p <0.001; I-HDRS elinganiselwe ngomtholampilo −7.2 ± 6.8, d = −1.03, CI [−1.41 −0.65]p < 1E-6) enezinzuzo ezihlala kuze kufike ekulandeleni kwamaviki angu-12. Ukukhathazeka okuhlangene nakho kwehliswe kakhulu (GAD7: d = -0.85, CI [−1.2, −0.49]p <0.001) kanye nezinga lempilo lithuthukisiwe (d = 0.68, CI [0.33, 1.02]p <0.001).
Ukwengeza, izinyathelo zokucabanga eziqondile ezithintekile ekucindezelekeni kufaka phakathi ukunaka okukhethiwe (d = 0.51, CI [0.18, 0.84]p <0.001), ukucubungula ukuphazamiseka (d = 0.53, CI [0.2, 0.85]p <0.01) nenkumbulo yokusebenza (d = 0.66, CI [0.31, 0.99]p <0.001) ibonise ukuthuthuka okubalulekile.
Ukulandelela kwe-EMA kuqinisekise ukuthi ukuthuthukiswa kwemizwa ecindezelekile kwabikezelwa ngokuqondile ngokuthuthuka kwezindlela zokuphila eziqondiswe ngabanye (β = 0.4 ± 0.09, p <0.0005). Okokugcina, ama-algorithms wesinqumo kanye nemodeli yolimi olukhulu (LLM) angafanisa isabelo se-iMAP esiholwa ngumqeqeshi ongumuntu ngokunemba okungafika ku-95%.
Uhlolo lwe-PerMA luveza indlela yokuphila yomuntu siqu yokungenelela yokucindezeleka kanye nokufaneleka ukukhuphuka nokuhlolwa kwe-RCT ukuze kutholwe ukusebenza kahle komtholampilo.
I-PERMA ibhaliswe ne-ClinicalTrials.gov ngaphansi kwenombolo yokubhalisa ethi NCT05662254.
