Discovery
- Hears about BioScent on social
- Visits the website
BioScent 博森
Globally, about 1 in 5 people are expected to develop cancer in their lifetime.
Yet many cancers begin changing the body long before symptoms become visible.
This led us to wonder whether the body might already be emitting subtle signals before symptoms appear.
Inspired by a dog's incredible sense of smell.
Research into canine scent detection revealed that dogs are capable of identifying cancers through VOCs emitted by the human body. If dogs can sense invisible biological changes, could we transfer that ability into technology? This became the starting point for exploring VOCs and air-based health detection systems.
Where we started — three hunches.
We didn’t treat these as facts — we turned each assumption into a research question, which led us into the science
Assumption 01
People don’t act on slight symptoms if it doesn’t affect their lives significantly.
Assumption 02
Certain illnesses doesn’t display visible early stage signs.
Assumption 03
They might be too busy hence doesn’t feel the need for minor health checks.
Assumptions → Research
We started with the assumption that the body may communicate early warning signs before symptoms become obvious. But rather than treating this as a fact, we used it as a research question: can biological changes be detected through signals the body naturally emits?
We analysed about 20+ papers on the following topics:
Relationship of VOC & Cancer
Insights
01
VOC changes can occur before visible symptoms
02
Breath contains detectable biomarkers
03
Existing detection methods are inaccessible
04
Pattern recognition matters more than single molecules
| Technology | Purpose | Limitation |
|---|---|---|
| Electronic Nose (eNose) | Uses sensor arrays to recognize VOC patterns in breath | Detects pattern rather than identifying exact molecules |
| GC-MS (Gas Chromatography-Mass Spectrometry) | Seperates and identifies VOC compounds in breath samples | Expensive and lab-dependent |
| Breath Biopsy Platforms | Collect and analyze exhaled breath biomarkers | Still emerging clinically |
| Nano Sensor Arrays | Nanomaterials detect chemical changes in breath | Requires further clinical validation |
The Science Behind Air-Based Cancer Detection
Key Research Insights
Insight 01
Cancer may alter human body’s VOC profile
Insight 02
Air-based detection already exists, but it’s lab-dependent
Insight 03
Personal baseline tracking could make detection more meaningful
Insight 04
Early detection is the real opportunity
So what did we do with these findings?
Opportunity 01
Make invisible health signals visible
Turn VOC shifts from breath, sweat and skin into simple, understandable health patterns.
Opportunity 02
Personalise detection through baseline tracking
Help users understand unusual changes based on their own normal VOC profile, not generic averages.
Opportunity 03
Design early alerts that inform, not alarm
Communicate uncertainty responsibly through calm, non-diagnostic alerts and guided next steps.
How might we help individuals recognise subtle biological changes earlier, while communicating uncertainty in a calm and trustworthy way?
Mapping where BioScent creates value.
To understand where BioScent could create value, we mapped the broader healthcare ecosystem. This helped us identify the stakeholders involved in detecting, interpreting, and acting upon health signals.
A day in the life of early awareness.
Primary user — busy professionals seeking earlier awareness of health changes through low-effort, non-invasive monitoring.
The journey
Education
- Reads ‘How it works’
- Checks the science behind it
- Compares with wearables
Onboarding
- Purchases device
- Downloads app, creates account
- First breath profile setup
Monitoring
- Passive VOC sensing
- Baseline building
Insight
- Receives a personalised health insight
- Sees an unusual change
Action
- Lifestyle adjustment
- Shares with caregiver
- Medical follow-up
Goal
- Informed action
- Continuous monitoring
Emotional arc
Key opportunities
Discovery
Clear value proposition
Education
Transparent evidence
Onboarding
Step-by-step setup guidance
Insight
Explain results in plain language
Action
Reassure & reduce anxiety
From a single breath to informed action.
1
Passive Sensing
BioScent continuously and passively senses VOCs in your everyday environment without disruption.
2
VOC Collection
Captures VOCs from breath, sweat, and skin to build a rich set of biological signals.
3
Baseline Learning
AI learns what is normal for each individual by understanding long-term patterns and context.
4
Pattern Shift Detection
Detects unusual changes or deviations from your personal baseline over time.
5
Personalised Health Insights
Complex data is translated into clear, actionable insights tailored to your unique health profile.
6
Informed Action
Empowers you and your support network to take informed action — from daily adjustments to professional care.
Continuous Learning Loop More data → Better baseline → More accurate insights → Better outcomes
Designed to inform, not alarm.
BioScent is a passive health-sensing system that combines VOC detection, AI baseline learning, and personalised insights to help users become aware of potential health changes earlier, without invasive testing.
Simple onboarding process
Guiding users through their first baseline capture without overwhelming them with technical detail.
Direct health index recognition
An immediate overview of their current health signal status. Clear health index states allows users to quickly understand whether they require further attention.
Plain language Science
Explains the science behind each signal in simple terms, helping users understand what changed without feeling overwhelmed
Personalised baseline over generic health scores
Measures 7 days health changes against the user’s personal baseline to highlight meaningful shifts over time.
What we'd need to solve before this is real.
Accuracy and false positives
The system would need clinical validation before being used for health decision-making.
Health anxiety
Alerts must be carefully designed to inform users without creating panic
Data privacy
Breath and body signal data would require strong consent, transparency and user control
Designing for health means designing for trust
What I learned
Health insights need to be designed with emotional safety, not just accuracy.
What was challenging
Translating complex VOC science into plain language without oversimplifying or making medical claims
My next steps
Test the prototype with users and healthcare professionals to validate trust, privacy and alert sensitivity
