Evidence-informed care
InclusiCare is grounded in research on caregiving, behavioral health, and health equity. Here's the evidence base that shapes our platform.
Core research foundations
These five papers directly inform InclusiCare's architecture — from CARLA's conversational extraction pipeline to our approach to behavioral data structuring and caregiver-facing AI design.
CaLM — Caregiving Language Model
Parmanto, B. et al. (2024). JMIR Formative Research, 8, e54633. DOI: 10.2196/54633
A direct precedent for CARLA. This study demonstrates that small, domain-specific language models using retrieval-augmented generation can outperform large general-purpose models for caregiving Q&A — validating InclusiCare's extraction-first, RAG-powered architecture.
Designing an AI Companion for Informal Caregivers
Walter, D. et al. (2025). BMC Nursing, 24, 1165. DOI: 10.1186/s12912-025-03868-2
Establishes seven empirically-derived design principles for AI-based caregiving companions, including continuous companionship, personalized assessment, and session memory. These principles directly mirror CARLA's design and support InclusiCare's cross-session continuity roadmap.
Structuring Clinical Speech Transcripts with Language Models
Corbeil, J.-P. et al. (2025). EMNLP Industry Track, pp. 859–870. DOI: 10.18653/v1/2025.emnlp-industry.58
Microsoft research that directly parallels CARLA's extraction pipeline — converting unstructured spoken language into structured clinical records. Validates InclusiCare's two-stage NLP extraction approach and introduces open-source benchmarks for clinical speech structuring.
Machine Learning for Cognitive Behavioral Analysis
Bhatt, P. et al. (2023). Brain Informatics, 10, 18. DOI: 10.1186/s40708-023-00196-6
A comprehensive review of ML approaches to behavioral analysis covering emotion detection, stress recognition, and abnormal behavior detection. Addresses the core challenges InclusiCare faces in working with behavioral data: multimodality, context-dependence, and annotation bias.
Generative AI for Assessment and Treatment of ASD — Scoping Review
(2025). Frontiers in Psychiatry, 16. DOI: 10.3389/fpsyt.2025.1628216
A systematic survey (2014–2025) of how generative AI is being applied to ASD screening, diagnosis, and intervention. Establishes the research landscape InclusiCare enters and highlights gaps — particularly around caregiver-facing tools and longitudinal behavioral data — that InclusiCare directly addresses.
Supporting research
Additional peer-reviewed work that broadens the evidence base for InclusiCare's approach to AI-assisted caregiving, behavioral data capture, and autism technology.
Breaking Barriers — AI and Assistive Technology in Autism Care
(2024). MDPI Healthcare. PMC open access
Reviews the full landscape of AI-integrated assistive technology in autism care — from robots and apps to NLP tools and wearables. Contextualizes InclusiCare's conversational logging approach within the broader ecosystem of tools available to families.
Machine Learning in ASD Assessment and Management
(2025). Pediatric Research. DOI: 10.1038/s41390-025-04566-0
Covers ML for ASD screening, phenotypic stratification, and personalized therapies. Provides strong context for InclusiCare's pattern detection roadmap and frames the platform within the broader ML-for-ASD research trajectory.
AI-Assisted Early Screening, Diagnosis, and Intervention for Autism in Young Children
(2025). Frontiers in Psychiatry, 16, 1513809. DOI: 10.3389/fpsyt.2025.1513809
Systematic review focused on AI tools for the youngest ASD population — infants, toddlers, and preschoolers. Directly relevant to InclusiCare's core user population and strengthens the case for early, continuous behavioral data capture.
Leveraging AI for Diagnosis and Treatment of ASD
Wankhede, N. et al. (2024). Asian Journal of Psychiatry, 101, 104241. DOI: 10.1016/j.ajp.2024.104241
Covers longitudinal monitoring, personalized treatment, and the challenges of data privacy and algorithmic bias in ASD AI tools. Maps directly to InclusiCare's product roadmap and provides useful framing for the platform's ethical data approach.
Evaluation Measures of LLMs for Family Caregiver Use
Han, S. et al. (2026). Digital Health. DOI: 10.1177/20552076261425343
A 2026 scoping review evaluating LLMs specifically for family caregiver use cases — assessing accuracy, reliability, readability, and comprehensiveness. Directly positions CARLA within the emerging evidence base for caregiver-facing language models.
Using LLMs to Identify High-Burden Informal Caregivers
Chien, S.-C. et al. (2024). Computer Methods and Programs in Biomedicine, 255, 108329. DOI: 10.1016/j.cmpb.2024.108329
Applied LLM use for caregiver burden identification in a real-world long-term care setting. Provides useful precedent for InclusiCare's potential population-level analytics and demonstrates clinical LLM applicability in caregiving contexts.
AI-Powered Chatbot for Early Detection of Caregiver Burden
Shankar, R. et al. (2025). Frontiers in Psychiatry, 16, 1553494. DOI: 10.3389/fpsyt.2025.1553494
Describes an AI chatbot specifically for early caregiver burden detection — the closest direct parallel to CARLA's conversational support function in the published literature. A strong reference point for InclusiCare's approach to detecting caregiver stress through natural conversation.
Exploring Needs and Challenges for AI in Nursing Care
(2023). BMC Digital Health. DOI: 10.1186/s44247-023-00015-2
A mixed-methods study identifying AI application scenarios in care settings, including the need for better tools to support informal caregivers and nursing assistants. Establishes that InclusiCare addresses a recognized gap in the care technology landscape.
Designing IT Applications for Informal Caregivers — Scoping Review
Premanandan, S. et al. (2024). Journal of Medical Internet Research, 26, e57393. DOI: 10.2196/57393
Comprehensive scoping review of design and evaluation principles for caregiver-facing IT applications. Provides a strong foundation for the UX and design rationale behind InclusiCare's interface choices and caregiver-first data authority model.
Artificial Intelligence Support for Informal Patient Caregivers
(2024). PMC. PMC open access
A focused review on AI tools specifically supporting informal (non-professional) caregivers — the exact population InclusiCare serves. Frames the platform within the growing evidence base for AI-powered caregiver support and validates the need for tools like CARLA.
Clinical partnerships
We're actively building relationships with clinical organizations to validate InclusiCare's impact on care outcomes, caregiver well-being, and health equity.
Partnership announcements coming soon. Interested in collaborating? Reach out.
Data methodology
InclusiCare captures behavioral and wellness data through natural-language conversations with CARLA, our AI care assistant. Data is processed using multi-provider AI models (with automatic fallback) and mapped to structured care profiles using FHIR R4-compliant schemas.
Privacy-first design
All care data is encrypted in transit and at rest. HIPAA-aligned practices. User-controlled data sharing.
Ethical AI
AI is used for extraction and pattern detection, not clinical decision-making. Human caregivers always have final say.
Interoperability
FHIR R4 compliance ensures data portability. Care data belongs to families, not platforms.
Outcomes
We're measuring InclusiCare's impact across several dimensions: caregiver confidence and burnout reduction, consistency of care across transitions, time saved on care documentation, and behavioral incident tracking accuracy.
Pilot outcome data will be published here as studies are completed.