MEMORI
An AI-enabled Clinical Decision Support tool seamlessly integrated with your hospital's EPR, empowering clinical teams to detect patient deteriorations earlier*.
Problem
Healthcare-associated conditions are on the rise, adding pressure to overstretched healthcare systems.
¹ NICE (2016). Healthcare-associated infections
Healthcare-associated infections Quality standard.
6-15%
of inpatients develop a healthcare-associated condition.⁽¹⁾
² Arefian, H., Hagel, S., Fischer, D., Scherag, A., Brunkhorst,
F.M., Maschmann, J. and Hartmann, M. (2019). Estimating
extra length of stay due to healthcare-associated infections
before and after implementation of a hospital-wide infection
control program. PLOS ONE, 14(5), p.e0217159.
³ Guest, J.F., Keating, T. and Gould, D. Modelling the annual
NHS costs and outcomes attributable to healthcare-associated
infections in England, BMJ Open 2020:10
8-12 days
in increased length of stay after a complication⁽²⁾, impacting patient flow across healthcare systems equating to 7.2-10.8M bed days.⁽³⁾
⁴ TheyWorkForYou. (n.d.). Hospital Beds: Costs.
£2.5-4.1b
in direct costs for the NHS arising from longer length of stay and cost of care.⁽⁴⁾
⁵ Schreiber, P.W., Sax, H., Wolfensberger, A., Clack, L. and
Kuster, S.P. (2018). The preventable proportion of
healthcare-associated infections 2005–2016: Systematic
review and meta-analysis. Infection Control & Hospital
Epidemiology, 39(11), pp.1277–1295.
35-55%
of infections are avoidable through earlier detection and intervention equating to 315,000-495,000 cases annually.⁽⁵⁾
Solution
MEMORI is a Clinical Decision Support tool designed to help clinical teams identify and understand emerging risks in patients
Using multimodal AI to analyse existing hospital data to predict patient risk
Continuous monitoring and risk stratification on general wards
Embedded in hospital EPR and workflows to support clinical decision-making
Predictive analytics leading to earlier intervention and improving clinical outcomes
AI-as-a-Medical Device*
*Pending regulatory approval as a Class IIb AI-as-a-Medical Device under EU:MDR
Features
1
Patient risk level integrated into existing EPR solutions and workflows
2
Easy identification of high-risk patients
3
Explainable AI to build trust and transparency in supported decision-making
4
Real-time updates and live data of patients' risk level
5
Next best actions based on risk and clinical guidelines
6
Key data trends in one place for efficient review
Integration
Integration with existing hospital EPR systems is key to support adoption and reduce cognitive burden on clinical teams.
MEMORI is system agnostic and seamlessly integrates into existing EPRs and health technology providers such as virtual wards or care management platforms.
Built around interoperability, MEMORI can process data from different interfaces and protocols, maximising its potential to support clinical decision-making.
Use cases
Infection risk prediction
MEMORI predicts the risk of a patient developing an infection within a given time period.
1
Patient identified as high risk for developing in nosocomial infection
2
Clinical team alerted via existing communication channels
3
Explainable AI offers rationale behind risk alert
4
Clinical Decision Support provided and indication of next best action
- Earlier intervention
- Better patient outcomes
- Improved hospital flow
- Reduced IV-antibiotic usage
- Better antimicrobial stewardship
- Reduced cost of care
Risk stratification in discharge, virtual wards, and remote patient monitoring
Discharge support to virtual ward
MEMORI facilitates discharge home and into the virtual ward.
1
Patient Identified as potentially suitable for discharge
2
MEMORI produces RAG risk score
3
Patient discharged to virtual ward with confidence
- Improved hospital flow
- Clinically: minimises clinical risk in the discharge process
- Promotes use of nurse-led/criteria-led discharge
Virtual ward & remote patient risk monitoring
Risk stratification in virtual ward/ remote patient monitoring pathways to highlight patients at risk of deterioration.
1
Patient enters relevant information
2
MEMORI calculates risk of deterioration
3
Clinician alerted if risk reaches pre-agreed threshold
4
Informed decision on next steps taken
- Reduces re-admission rates
- Priorities workflows
- Promotes proactive & preventative healthcare
Safety, compliance and AI-bias detection
MEMORI is to be regulated as a class IIb AI-as-a-Medical Device (currently under audit by Scarlet Comply).
In addition, Sanome adheres to the latest healthcare safety and quality standards such as DSPT, ISO14971:2019, EN 62304: 2006, EN82304-1:2016, ISO13485 and IEC 62366 and GDPR.
Sanome has a rigorous testing framework for our machine learning models (MLOps) which includes tests for architectural robustness, adversarial robustness, domain shift, and explainability. Robust assessments help ensure our models minimise any AI bias.
FAQs
How can I learn more about MEMORI?
Contact us – we can provide you with more information on our progress, and how you can be part of the journey.
Is MEMORI deployed on premise or in the cloud?
We offer flexible deployment options to suit your organisation’s needs. Our solutions can be deployed in the cloud, providing scalability and accessibility, or on-premises for those requiring full control over infrastructure and data security.
What EPRs can your solution integrate with?
We can integrate seamlessly with any EPR that uses common interoperability standards such as HL7v2, FHIR.
How do you keep your platform secure?
Our platform uses robust encryption protocols and role-based access control to ensure the highest levels of data security. Our platform is designed with GDPR principles embedded from the outset, prioritising privacy and compliance. We assess and enhance our security measures on a regular basis to monitor emerging threats. Additionally, the company holds certifications such as DSPT and Cyber Essentials.
How does Sanome minimise AI bias?
Sanome has a rigorous testing framework for our machine learning models (MLOps) which includes tests for architectural robustness, adversarial robustness, domain shift, and explainability. Robust assessments help ensure our models minimise any AI bias.
How does MEMORI impact the NHS workforce?
We strive to empower clinicians by supporting clinical decision making. This decreases cognitive burden and allows teams to reinvest that time into patient care.
Are there any financial benefits to Trusts?
The prevention of patient deterioration positively impacts a hospital’s average length of stay, and thus bed capacity. This presents the opportunity to increase elective activity, get Trusts closer to the 107% needed to access the Elective Recovery Fund, and generate the income needed to reinvest into their services.
How does MEMORI support population health?
The current demand for health services drastically outweighs the system’s capacity. Earlier detection and prevention of deterioration keeps patients out of hospital and decreases the care burden.
What's MEMORI's impact on health inequalities?
MEMORI seeks to positively contribute to the eradication of health inequalities, as we aim to champion prevention and mitigate against digital exclusion by delivering cutting-edge AI to all, regardless of their digital literacy.
Safety
NHS Digital Data Security and Protection Toolkit
Secure role-based access via EPR SSO
Patient consent (as required)
On prem or cloud set up – installed by Sanome team
All data encrypted