Automated conversational artificial intelligence (AI) for outpatient malignant bowel obstruction (MBO) symptom monitoring.
Notice bibliographique
Résumé
1547 Background: MBO is a severe complication of advanced cancer. A Canadian ambulatory MBO program with nurse-led proactive call management demonstrated reduced hospitalization rates and improved survival. To overcome resource limitations, a smartphone app was developed, achieving 65% adherence. Building on this foundation, automated phone calls offer a promising approach to enhance adherence and improve symptom monitoring. Methods: We conducted a prospective pilot study at a tertiary Spanish hospital to remotely monitor MBO signs and symptoms using a conversational AI-based platform (Lola-Tucuvi). Patients (pts) with cancer with an active MBO or at risk of developing it (per PMMBO criteria) were enrolled. Automated, interactive phone calls were performed by the platform (Lola) weekly or biweekly. Lola performed structured MBO symptom assessments utilizing advanced natural language processing and AI algorithms, to analyze responses in real time. Alerts were generated for moderate or severe symptoms, which were flagged on a dashboard. Nurses contacted pts based on alerts. The primary objective was feasibility measured by adherence (% of answered calls), with a hypothesized adherence of ≥65% considered optimal. Results: From January 2024 to January 2025, 54 pts were enrolled, with 25 still active at the time of analysis. Median age was 60 years (range 29-86), and 96% of pts are female. Type of tumors included gynecologic (87%) and gastrointestinal (13%). All pts were on systemic therapy: chemotherapy (50%), immunotherapy (24%), ADC (15%), targeted (11%). Median prior lines of therapy were 2 (1-6), and 41% (22/54) of pts had an active MBO prior to enrollment. Lola performed 716 phone calls and 645 were answered, with an adherence of 90%. This resulted in an estimated 183.2 hours of nursing call time saved. Median time on the program was 117 days (7-356), and pts received a median of 14 calls. Of answered calls, the 36% (234/645) generated alerts, with 44% classified as severe. Most frequent severe and moderate alerts were constipation and abdominal pain, respectively. Nurses acted on 73% (171/234) of the alerts, providing interventions such as dietary modifications, medication adjustments, clinical or emergency assessments. During follow-up in the program 31.5% (17/54) of pts had ≥1 active MBO and 18.5% (10/54) required admissions for MBO. Feedback was received from 26 pts, indicating a high satisfaction (4.6/5), and 96% would recommend the use of Lola. Conclusions: This conversational AI platform demonstrated excellent feasibility with 90% adherence, higher than prior app-based solutions. It effectively monitored MBO symptoms, enabling timely clinical interventions and enhancing patient engagement. These results highlight the potential of AI-driven remote monitoring system to improve outcomes in cancer care. Further validation through randomized studies is warranted.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,001 | 0,005 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,001 | 0,000 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,001 | 0,000 |
| Science ouverte | 0,000 | 0,001 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,002 | 0,001 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».