AI-Based Cognitive Gaming and Memory Assistance Platform for Elderly Dementia Patients in North Eastern Region (NER)
Ministry of Development of North Eastern Region (MDoNER) · Space Technology · Software
The safest high-ceiling pick in the MDoNER block, but the description asks for five components rather than a games app, so budget for the reminder and caregiver halves from the start.
What it actually is
Elderly people with dementia in the Northeast have almost no access to cognitive therapy, and their families carry the whole load without guidance. The ask is an app with brain-training games plus everyday memory support, in languages and imagery the patient recognises. A caregiver should be able to see whether the person is getting better or worse over time.
What to build
A tablet-first app with a patient mode containing adaptive cognitive games across the four domains the description names — memory, attention and concentration, daily routine recall, and pattern and object recognition — with difficulty that auto-adjusts to the patient's performance, a memory-assist screen carrying the reminder categories specified (medicines, hydration, daily activities, medical appointments), voice-assisted multilingual interaction with culturally familiar NER visuals and sounds, full offline operation with sync when connectivity returns, and a separate caregiver dashboard plotting per-domain cognitive scores over time with a decline alert.
Smallest thing that wins the room
Play two rounds of the recall game with deliberately wrong answers, watch the difficulty drop live, then flip to the caregiver dashboard and show the memory-domain line bending downward with a flagged alert.
How crowded this one gets
A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.
Quieter than 68% of the 226 · #73 of 226 by expected field
A normal-sized field. Your idea has to be good, not miraculous.
Why: central ministry statements sat below the average.
This is a guess, not a fact
Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.
The scores
The number is the shorthand. The line under it is the reason.
Acceptance potential
5/5A clear hidden gem — pure software with zero data risk, a hands-on demo, obvious human impact, and the portal has filed it under Space Technology, so teams browsing MedTech or Smart Education will never find it.
Feasibility
5/5There is no model to train, no dataset to source and no hardware to buy — the description asks only for adaptive game logic, reminders, a voice interface and a dashboard, all of which are application-layer work.
Innovation scope
3/5The description prescribes the game domains, the reminder categories, the offline requirement and the caregiver dashboard, so the shape is largely fixed and your room is confined to the adaptive difficulty policy and the cultural localisation.
Clarity
5/5The description enumerates the cognitive domains, the reminder types, the multilingual and offline requirements and the caregiver monitoring separately, leaving almost nothing about the deliverable to interpretation.
Effort
HeavyFour game domains, a reminder system, a voice-enabled multilingual interface, offline sync and a caregiver analytics dashboard is a lot of separate surface area even though no single piece is technically deep.
Demo-ability
EasyA judge can pick up the tablet and play the game themselves within thirty seconds, which is the strongest demo format available and almost no other PS in this ministry offers it.
In its favour
- Green flag: The theme mislabelling is a genuine competitive advantage — anyone filtering for MedTech or Smart Education will not see this PS at all
- Green flag: The description explicitly requires offline functionality, which means the rural connectivity question every judge asks is already answered by the spec rather than by you improvising
- Green flag: Zero dataset dependency, so the project cannot collapse because a data source fell through
- Green flag: The four cognitive domains are named in the description, so your scope is defined for you and cannot be criticised as arbitrary
Against it
- Red flag: Evidence that brain-training slows dementia progression is genuinely contested, and a clinician judge will ask what your games are grounded in — have MoCA or ADAS-Cog domain mapping ready
- Red flag: Elderly UI is a real discipline most student teams fail: small fonts, low contrast, swipe gestures and timers are all disqualifying for this user
- Red flag: The NER framing plus the multilingual requirement implies Assamese, Manipuri, Khasi and Mizo support, which is substantial content work teams discover late
- Red flag: The description asks for games, reminders, voice, offline sync and caregiver analytics — that is five components, and a team that polishes the games and stubs the rest has answered a fraction of the statement
What you will be writing
- Flutter or React Native
- Bhashini API for Indic TTS/STT
- SQLite / Hive offline store with sync
- MediaPipe face detection for family album
- fl_chart caregiver analytics
- MoCA/MMSE-aligned domain scoring
- Geriatric care
- Cognitive health
- Assistive technology
Prior art to read before you start
cognitive training games for dementia · caregiver monitoring dashboard · multilingual elderly assistive app
Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.