Patient deteriorations, detected sooner.
The early signs of infection are often already in the patient record. Sanome brings them together, so clinical teams can see who is at risk, and why, in time to act. Good for patients. Good for hospital finances.
17.8h
earlier warning¹
39%
more infections detected¹
Class IIb
CE-marked medical device
Working with NHS Trusts and EPR suppliers
The warning signs are already in the data.
They build quietly across vitals, bloods and notes, and are often seen too late.
Meet MEMORI.
MEMORI is certified AI that predicts hospital-acquired infection, starting where deterioration is most common.
It runs inside the EPR your team already uses, including Epic, Altera Sunrise and PatientSource, with no new system, hardware or log-in. Every alert shows the reasons behind it and the next best action.
Earlier, better, safer.
Real-world evidence from live use at the Royal Hospital for Neuro-disability.
EARLIER
earlier warning than NEWS2¹
Mean lead time, n=27 paired events, p=0.0008.
BETTER
more infections detected¹
Qualified infections, n=71 events, p=0.003.
SAFER
cases NEWS2 caught that MEMORI missed¹
7 events NEWS2 missed that MEMORI caught; none the other way.
“
MEMORI helps nurses say out loud, could this be sepsis, far better than NEWS2 ever could.
Matron
Royal Hospital for Neuro-disability
Value for hospitals, and the people in them.
Released capacity, sustained use, and a return you can model for your own beds.
TRUSTS
a year released for elective recovery⁵
Modelled for 614 monitored beds: 1,559 bed days a year.
CLINICIANS
sustained clinical adoption⁶
Up from 67% over five months (n=65).
IT TEAMS
uptime, inside the EPR⁶
Added through a Change Control Notice.
In use on the ward today.
Photo: Mark Norman / BBC
BBC NEWS · 7 JULY 2026
Hospital first to use AI for spotting infections.
East Kent became the first NHS trust to use MEMORI, at Kent and Canterbury Hospital.
Earning the trust of patients and clinicians.
Regulated, secure and monitored for safety, cloud-hosted or on-premise.
Benedikt von Thüngen
Founder and CEO, Sanome
Bringing clinical AI to the bedside.
Healthcare has thousands of clinical AI models. Few reach patients, because three hurdles stand in the way. We have cleared all three.
Regulation
CE-marked Class IIb under EU MDR.
EPR integration
Live inside the systems hospitals already run.
Clinical trust
Explainable, and used by clinicians every day.
“
This is just the beginning. MEMORI is there to predict hospital-acquired infections, but in the future we want to use this collaboration to do so much more.
Dr Mike Bedford
East Kent Hospitals University NHS Foundation Trust
1. Sanome, MEMORI HAI monthly outcome analysis, September 2026. Data on file.
2. Eurostat. Preventable and treatable mortality statistics. Statistics Explained, 2025 (2023 data).
3. Guest JF, Keating T, Gould D, Wigglesworth N. Modelling the annual NHS costs and outcomes attributable to healthcare-associated infections in England. BMJ Open. 2020;10:e033367.
4. World Economic Forum. 4 ways data is improving healthcare. December 2019.
5. Sanome, MEMORI HAI ROI Calculator, Internal analysis tool, October 2026.
6. Sanome, MEMORI HAI PMS/PMCF Report, Month 6, August 2026. Data on file.