Discens Machina builds AI systems that run inside your infrastructure. They repair DTS coding, catch errors in discharge documents, and turn speech into structured reports — without sending data outside the hospital.
Piloted at KBC Split. Built in collaboration with clinicians.
Doctors are overloaded. Documentation is imperfect. A system that ignores this doesn't work. A system that helps must fit inside the existing workflow, not beside it.
Each system solves a specific problem. They share infrastructure, security guarantees, and a single integration with the hospital information system.
Incomplete diagnosis capture leads to incorrect DTS grouping. Up to 10% of hospital revenue can be lost due to undercoding and missing diagnoses.
Diagnoses are not fully documented
Chronic conditions are omitted
Supporting information is not captured
Hospitals recover revenue currently lost due to incomplete documentation.
Errors in discharge documentation can directly impact patient outcomes. Most are not random — they cluster at the same handover points.
Inconsistencies between diagnosis and therapy
Missing critical information
Unclear discharge instructions
Improved patient safety and more reliable documentation.
Doctors spend significant time on documentation — time that doesn't reach the bedside. The fix is structural: capture what they say, in the structure the report needs.
More time for patient care and reduced workload.
The fixes are not new vendors. They are new defaults — inside the same hospital, on the same data.
Everything executes inside the hospital. Data never leaves your perimeter.
Models tuned on real medical documentation in Croatian.
The founder is an MD. Every system is validated in clinical reality.
Works with your existing system. No rip-and-replace.
Discens Machina installs on your infrastructure. No external cloud, no data leaving the perimeter, no copies held by us. What enters the system stays in the system — under your control, your policies, your audit.
Hospital network
│
├── HIS · RIS · LIS
│
└── Discens Machina (on-prem)
├── DTS assistant
├── Clinipal
└── Transcription
→ No data leaves the perimeter.Installed on existing infrastructure, typically within a week.
HL7/FHIR connectors; adapts to your HIS without replacement.
GDPR and ISO 27001 posture; audit trail for every query.
First measurable results within the first month of operation.
The founder is a physician with clinical experience in Croatian public healthcare. Every system emerged from a problem seen on the ward — not from a slide deck.
Discens Machina is a startup spin-off from the University of Split School of Medicine.
— Discens Machina
Exclusively inside the hospital's infrastructure. No external cloud, no transfer across the perimeter, no copies held by us.
The demo takes 30 minutes. We show real cases from the pilot, not marketing footage. After the demo you decide whether the conversation is worth continuing.