{"id":1181796,"date":"2026-08-16T12:12:54","date_gmt":"2026-08-16T19:12:54","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/utility-of-an-llm-powered-experts-in-the-loop-chatbot-for-pre-and-post-operative-care-of-cataract-surgery-patients\/"},"modified":"2026-08-20T15:47:05","modified_gmt":"2026-08-20T22:47:05","slug":"utility-of-an-llm-powered-experts-in-the-loop-chatbot-for-pre-and-post-operative-care-of-cataract-surgery-patients","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/utility-of-an-llm-powered-experts-in-the-loop-chatbot-for-pre-and-post-operative-care-of-cataract-surgery-patients\/","title":{"rendered":"Utility of an LLM-powered experts-in-the-loop chatbot for pre- and post-operative care of cataract surgery patients"},"content":{"rendered":"\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Purpose To evaluate the utility of CataractBot, an LLM (Large Language Model)-powered chatbot that provides doctor-verified answers to patient questions about cataract surgery. We examine its use by both end-users (patients and attendants) and medical experts. Methods A 24-week study was conducted to evaluate CataractBot among patients, their attendants, doctors, and patient coordinators. The bot responded instantly to questions by querying a knowledge base curated by medical professionals. Each response was asynchronously verified by an ophthalmologist (for medical questions) or a patient coordinator (for logistical questions), and their edits contributed to updating the knowledge base, thereby minimizing future expert intervention. A mixed-methods analysis was conducted on interaction logs, including patient and attendant questions, chatbot answers, and expert verifications. Results A total of 318 patients and attendants sent 1,992 messages, and LLM-generated answers were verified by five doctors and two coordinators. Questions asked pre-surgery were significantly more than post-surgery ( p < 0.001 ) . Participants asked significantly more medical than logistical questions ( t 309 = 7.3 , p < 0.001 ) . Doctors rated 84.5% of CataractBot\u2019s answers to medical questions as accurate and complete. Their edits, which mainly involved adding information, increased the acceptance of the bot\u2019s answers by 19.0% over time. Conclusion CataractBot was predominantly used to address medical questions. It incorporated expert corrections to improve its answers and reduce the experts\u2019 bot-related workload over time. This study highlights the potential of LLM-powered chatbots to support patient-provider communication in ophthalmology.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Purpose To evaluate the utility of CataractBot, an LLM (Large Language Model)-powered chatbot that provides doctor-verified answers to patient questions about cataract surgery. We examine its use by both end-users (patients and attendants) and medical experts. Methods A 24-week study was conducted to evaluate CataractBot among patients, their attendants, doctors, and patient coordinators. The bot [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":[{"type":"text","value":"Bhuvan Sachdeva","user_id":0},{"type":"user_nicename","value":"Pragnya Ramjee","user_id":"43590"},{"type":"user_nicename","value":"Rahul Sharma","user_id":"36308"},{"type":"text","value":"Mithun Thulasidas","user_id":0},{"type":"text","value":"Sowmya Raveendra Murthy","user_id":0},{"type":"text","value":"Geeta Fulari","user_id":0},{"type":"text","value":"Kaushik Murali","user_id":0},{"type":"user_nicename","value":"Mohit Jain","user_id":"38769"}],"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"European Journal of Ophthalmology","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"European Journal of Ophthalmology","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"561","msr_page_range_end":"568","msr_series":"","msr_volume":"36","msr_copyright":"","msr_conference_name":"","msr_doi":"10.1177\/11206721251396664","msr_arxiv_id":"","msr_mag_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_release_tracker_id":"","msr_highlight_type":"","msr_date_display_format":"","msr_main_download_label":"","msr_external_link_label":"","msr_doi_label":"","msr_published_date":"2025-12-08","msr_startdate":"","msr_presentation_date":"","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_year":2025,"msr_month":12,"msr_day":8,"msr_microsoftintellectualproperty":false,"msr_pub_id":"8122da230214b3ad5d45935a9298bd5d81ff144a","msr_publication_uploader":[{"type":"url","title":"https:\/\/pubmed.ncbi.nlm.nih.gov\/41358933","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_original_fields_of_study":[],"msr_s2_paper_id":"8122da230214b3ad5d45935a9298bd5d81ff144a","msr_s2_pdf_url":"","msr_citation_count_updated":"","msr_citation_count":2,"msr_influential_citations":0,"msr_reference_count":24,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[{"provider":"s2","id":"8122da230214b3ad5d45935a9298bd5d81ff144a"},{"provider":"doi","id":"10.1177\/11206721251396664"},{"provider":"pubmed","id":"41358933"},{"provider":"corpusid","id":"283688704"}],"msr_hide_image_in_river":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13556,13553],"msr-publication-type":[193715],"msr-publisher":[],"msr-publication-cta":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[261083,246985],"msr-conference":[],"msr-journal":[270519],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-1181796","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-medical-health-genomics","msr-locale-en_us","msr-field-of-study-digital-health","msr-field-of-study-medicine"],"msr_publishername":"","msr_edition":"European Journal of Ophthalmology","msr_affiliation":"","msr_published_date":"2025-12-08","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"European Journal of Ophthalmology","msr_volume":"36","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":0,"msr_main_download":"","msr_publicationurl":"","msr_doi":"10.1177\/11206721251396664","msr_publication_uploader":[{"type":"url","title":"https:\/\/pubmed.ncbi.nlm.nih.gov\/41358933","label_id":243109,"id":false,"viewUrl":false}],"msr_related_uploader":[],"msr_citation_count":2,"msr_citation_count_updated":"","msr_s2_paper_id":"8122da230214b3ad5d45935a9298bd5d81ff144a","msr_influential_citations":0,"msr_reference_count":24,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Bhuvan Sachdeva","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Pragnya Ramjee","user_id":43590,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Pragnya Ramjee"},{"type":"user_nicename","value":"Rahul Sharma","user_id":36308,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Rahul Sharma"},{"type":"text","value":"Mithun Thulasidas","user_id":0,"rest_url":false},{"type":"text","value":"Sowmya Raveendra Murthy","user_id":0,"rest_url":false},{"type":"text","value":"Geeta Fulari","user_id":0,"rest_url":false},{"type":"text","value":"Kaushik Murali","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Mohit Jain","user_id":38769,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Mohit Jain"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"article","related_content":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1181796","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":2,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1181796\/revisions"}],"predecessor-version":[{"id":1184419,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/1181796\/revisions\/1184419"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1181796"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=1181796"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1181796"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=1181796"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=1181796"},{"taxonomy":"msr-publication-cta","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-cta?post=1181796"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=1181796"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1181796"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1181796"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=1181796"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=1181796"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=1181796"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1181796"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=1181796"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}