Healthy Longevity with
Dignity and Identity
Reteena is an applied ai lab focused on creating human-friendly
solutions that enable more accessible alzheimer's diagnosis and therapy
01
Remembrance is an emotionally intelligent AI memory companion for Alzheimer's patients that delivers reminiscence therapy through natural conversations. By storing memories in interconnected knowledge graphs that mirror how human memory actually works, Remembrance helps patients maintain their sense of identity and connection to their life stories even as cognitive decline progresses.
02
Enhances low-field MRI scans for Alzheimer’s diagnosis using deep learning-based brain segmentation and volumetric analysis. By enabling 96% diagnostic accuracy on affordable, low-strength scanners, the framework provides a faster, cost-effective alternative to conventional high-field MRI while preserving clinically meaningful structural information for early and accessible neurodegenerative assessment.
03
Lark is a biologically inspired decision-making framework that combines LLM-based reasoning with an evolutionary, stakeholder-aware multi-agent system. Through plasticity, duplication and maturation, ranked-choice aggregation, and compute-aware optimization, Lark generates concise, aligned strategies while balancing stakeholder trade-offs, performance, and computational cost across decision cycles.
04
GeneAttentionNet is an interpretable, biologically informed neural architecture for Alzheimer’s classification from gene expression data. Using attention-based tokenization, pathway gating, and protein-interaction–masked heads, it captures gene-level context, improving accuracy over baselines while enabling trustworthy, mechanism-aware predictions for transcriptomic disease modeling and biomedical research.
05
CortiForge is an open-source, biologically grounded framework for modeling cortical microcircuits that links spiking neural network simulations with experimental brain data. Through standardized neuron libraries, AllenSDK integration, and end-to-end workflows, CortiForge enables interpretable, validated circuit models that bridge theory, experiments, and neuroscience research at scale reproducibly.
