Industry
Healthcare
Patient-centered Care: Health and Wellness Redefined
Highlights
- Multiple sclerosis (MS) is a complex, evolving condition that demands personalised treatment, as traditional static monitoring fails to capture real-time, subjective symptoms.
- By leveraging wearables, smartphones, and non-invasive sensors, specific, quantitative indicators can be gathered to provide a comprehensive, real-time view of the patient’s functional health.
- Multiple sclerosis is a lifelong condition without any cure, but its symptoms can be effectively monitored as well as managed for better well-being.
On this page
On this pageinpage
Lost data
Multiple sclerosis (MS) is a chronic condition that affects the central nervous system, disrupting communication between the brain and the body.
MS is often called the 'disease of a thousand faces' because its symptoms vary widely not only from person to person but even on a day-to-day basis. Common symptoms include fatigue, blurred vision, muscle stiffness, loss of balance, and cognitive challenges. These symptoms are difficult to identify externally, unpredictable, change rapidly, and can be attributed to many other causes besides MS. All these make MS especially difficult to manage.
Such transient and subjective symptoms can occur multiple times in a single day, making it difficult to track and report accurately. Often:
- Patients are too fatigued or overwhelmed to log every detail.
- Symptoms seem minor or temporary, so they are forgotten by the time patients schedule their subsequent doctor appointment.
- Doctors only see a snapshot, missing the real-world context.
This creates a clinical blind spot. Without continuous data, treatment decisions rely on incomplete information, increasing the risk of relapses or progression.
Personalisation
Symptoms of MS are highly subjective as it is governed by the respective immune systems of affected individuals.
Even when two patients have lesions in the same area of the brain or spinal cord, their symptoms can still differ dramatically. For example, one patient may experience vision problems, while another struggles with balance or speech. The difference lies in individual brain connectivity, immune response, and overall health profile.
This means treatment cannot be one size fits all. It must be personalised, considering both the patients’ medical history and their unique symptom patterns.
Bridging the gap
Modern healthcare is moving towards integrated monitoring systems that combine the depth of past medical records with the immediacy of real-world data.
Historical information from magnetic resonance imaging (MRI) scans, lab results, and neurological tests, provides a strong foundation for understanding disease history and progression. But MS is dynamic, and what happens between clinic visits often goes unseen. That’s where digital biomarkers come in.
Using wearables, smartphones, and non-invasive sensors, a holistic view of the patient’s functional reality can be captured through specific, quantitative indicators:
- Gait and mobility: Motion sensors embedded in smartwatches, intertial measurement units (IMUs) in the shoe and smartphone, pressure-sensitive gait pads placed in shoes or insoles autonomously track walking speed, stride length, and balance variability. Advanced gait pads measure pressure distribution, step symmetry, and weight transfer, offering granular insights into mobility changes. These metrics help identify subtle, pre-symptomatic deterioration, often before the patient notices it. Combined with accelerometer and gyroscope data from wearables, clinicians can detect early signs of motor impairment and intervene proactively.
- Speech and cognitive function: Acoustic analysis monitors vocal tone, pause duration, and speech clarity to detect early signs of cognitive fatigue or motor control issues. To ensure privacy, speech data is processed on device, extracting only numerical features (such as pause ratios and articulation rate) without storing raw audio. This means sensitive voice recordings never leave the patient’s device. Passive tracking of keystroke dynamics such as typing speed and error patterns adds another layer, revealing changes in information processing speed without capturing actual text content, preserving confidentiality.
- Physiological metrics: Continuous monitoring of heart rate and heart rate variability (HRV) gives insights into autonomic nervous system health. Additional sensors measure stress levels, skin temperature, and sleep quality, correlating these factors directly with symptom severity and disease activity. For example, sustained HRV suppression combined with poor sleep efficiency can indicate heightened fatigue or early relapse risk.
- Vision changes: Smart glasses or eyelid-mounted sensors can monitor blink rate, eye closure duration, and saccadic movements to detect visual fatigue and optic nerve involvement. Infrared sensors embedded in eyewear measure pupillary light reflex, providing early indicators of optic neuritis. Data is processed locally and securely transmitted for comparison against personal baselines.
- Muscle stiffness: Surface electromyography (sEMG) patches or smart clothing capture muscle activation patterns and tone, identifying spasticity or stiffness. IMUs complement this by detecting restricted joint range and rigidity during movement. Continuous sEMG readings combined with motion data create a stiffness index for proactive alerts.
- Cognition and fatigue: Lightweight EEG headbands track brainwave patterns related to attention and processing speed, while passive metrics such as keystroke dynamics and touchscreen interactions reveal subtle cognitive slowing. AI models integrate EEG signals with behavioural data to identify sustained cognitive decline beyond normal fatigue. Similarly, eyelid-mounted magnetoelastic sensors assess fatigue through blink frequency and closure time, while HRV and skin temperature provide physiological insights. Multi-sensor fusion combines these indicators to distinguish MS-related fatigue from routine tiredness. By linking these real-time digital signals with historical clinical data, a complete, personalised health profile can be created enabling proactive care and timely interventions.
MS patients can comfortably track key indicators like gait, speech patterns, heart rate variability (HRV), and sleep passively through a smartwatch and smartphone, which integrate seamlessly into daily life without adding strain. More specialised sensors such as EEG headbands, smart glasses, and sEMG patches will be reserved for short, scheduled assessments triggered by specific symptoms or clinical needs. This approach minimises patient burden while ensuring comprehensive monitoring and timely interventions when necessary.
Personal baseline
To understand what is normal for a person, the system passively monitors for a period by tracking gait and mobility, speech and cognitive function, heart and sleep metrics, among other parameters, and builds personalised baseline.
This baseline is personalised, not generic. It reflects each patient’s unique lifestyle and health patterns.
Intelligent alerts: After establishing the baseline, the system continuously tracks and compares real-time data against this personalised profile. If it detects persistent deviations—such as slower walking speed, increased speech pauses, or reduced sleep quality, it compares these signals against the patient’s baseline and clinical history.
When patterns suggest disease activity or treatment drift, the system generates smart alerts. These alerts can:
- Prompt the patient to consult their doctor for further evaluation.
- Highlight contributing factors, such as poor sleep or elevated stress, so patients can take corrective steps.
Smart alerts with reason codes include context such as: ‘Mobility decline beyond normal fatigue pattern’; ‘speech slowdown persists despite adequate sleep, possible MS activity’, or ‘sleep disruption likely driving fatigue, consider lifestyle adjustment’.
Layered approach
Not every symptom is MS-related; to avoid false alarms, the system uses multi-layered intelligence validated by research.
Symptoms such as fatigue after a long day or even slower typing caused by stress can mimic MS signs. The following multi-layered approach can help diagnose MS:
- Personalised baseline modeling: The system models each person’s ‘normal’ by time of day, activity level, and lifestyle. Digital phenotyping studies emphasise within-person baselines to separate routine variability from clinically meaningful change.
- Contextual data fusion: Multi-sensor fusion (gait, speech, HRV, sleep) improves reliability and reduces false positives; fusion frameworks in healthcare (Bayesian, machine learning (ML)) are shown to enhance decision quality.
- Trend analysis (vs single events): Longitudinal models detect sustained changes rather than isolated spikes, aligning with research showing trend-based monitoring captures clinically relevant deterioration.
- Clinical anchors: Signals are cross-checked with MRI activity, serum neurofilament light chain (sNfL), and optical coherence tomography (OCT) biomarkers that are linked with future relapses and progression, consequently alerts are evidence-linked, not just sensor-driven.
Systematic reviews and prospective studies support this layered approach as both feasible and clinically meaningful for motor, cognitive, mood, sleep, and fatigue monitoring.
Actionable insights
Collecting data is only the first step. What truly matters is how it helps doctors make better decisions.
- Unified dashboard: All patient data, wearable metrics, symptom trends, and clinical records will be securely consolidated into a cloud-based dashboard, integrating tracked metrics and clinical records in one place, and highlighting deviations along with potential contributing factors such as reduced sleep, increased workload, or other lifestyle changes.
- Visual health timeline: Doctors can view a timeline of symptom patterns, highlighting deviations from the patient’s baseline and correlating them with MRI findings or lab results.
- Artificial intelligence (AI)-powered summaries: Advanced analytics generate summary reports before each appointment, flagging potential disease activity, treatment drift, or lifestyle factors impacting symptoms.
Conclusion
This approach is a big step forward in managing multiple sclerosis. It makes life easier for patients by removing the need to track symptoms manually, gives clinicians a clear, real-time picture of health, and helps improve outcomes through timely, personalised care.
By integrating historical medical records, personalised baselines, and real-world digital biomarkers, healthcare can transition from reactive treatment to proactive, precision-driven care. This digital AI-powered approach makes invisible symptoms visible and actionable, creating a smarter, connected, and patient-centric ecosystem that not only transforms MS management but also sets the stage for addressing other complex chronic conditions in the future.