ZEKLAR
EN / IT
← Work
KAIROS · MedTech · 2025

KAIROS

The digital twin for safer dosing

KAIROS is an in-silico patient digital twin: it fuses genomic profile, physiology and drug to estimate exposure, toxicity and efficacy before administration. Change the dose or a variant and watch risk and efficacy shift in real time, decision support for oncology dosing. A research tool (not a medical device).

Year

2025

Client

Confidential clinical research consortium, Northern Italy

Role

System architecture, platform development, HPC integration

Duration

11 months

Team

5 engineers, 1 biomedical researcher, 1 designer

Category

Bio-Engineering / MedTech

// Results

92%

Concordance (in-silico)

−78%

CFD time reduction

0.91

Segmentation Dice

340

Cohort (simulated)

<4 h

End-to-end sim time

2,400

Annotated scans

// The context

Every year, thousands of oncology patients endure chemotherapy regimens calibrated through trial and error. Miscalibrated doses cause severe toxicity or under-treatment. High-resolution genomic sequencing and MRI scans already exist in abundance, yet they sit in separate silos, never unified into a predictive model. The field needed an infrastructure capable of bridging molecular biology, anatomy, and pharmacokinetics in a single real-time pipeline.

// The challenge

The real obstacle wasn't raw compute. It was data coherence: making 150 GB whole-genome sequencing files, 3D vascular meshes extracted from DICOM, and pharmacokinetic models speak to one another on a single computational graph without sacrificing scientific fidelity. Initial CFD simulation times exceeded 18 hours per patient, making clinical use a non-starter. Each rare genomic variant demanded its own validation pathway.

// The solution

Zeklar designed a directed-graph Nextflow pipeline that orchestrates genomic ingestion, PyTorch3D vascular segmentation, and the CUDA CFD solver in parallel stages, cutting end-to-end time to under four hours on an A100 cluster. For vascular meshes we trained a 3D segmentation model on 2,400 radiologist-annotated scans, reaching a Dice score of 0.91. The pharmacokinetic engine maps patient-specific SNP and CNV variants onto a Starling-Onsager model calibrated per patient. NVIDIA Clara provided the DICOM-to-tensor abstraction that made KAIROS compatible with existing hospital PACS.

// What made it special
01

Patient-specific digital twin

Every simulation fuses WGS data, 3D vascular morphology, and individual physiological parameters into a single coherent computational graph.

02

Vascular CFD at clinical speed

Multi-GPU CUDA optimisation cut CFD runtimes from 18 hours to under four, bringing the platform within a real oncology ward round.

03

High-fidelity segmentation

The PyTorch3D model extracts vascular meshes at a Dice score of 0.91, outperforming published benchmarks by 7 points.

04

92% in-silico concordance

Measured on a SIMULATED 340-case validation cohort. Not yet validated on real clinical outcomes. Research Use Only.

// The process
01

Discovery & data modelling

We mapped clinical data flows, defined the unified genomic-physiological schema, and locked validation criteria with the research team.

02

Segmentation pipeline

We trained the 3D MRI segmentation model, integrated NVIDIA Clara, and automated vascular mesh production end to end.

03

CFD & pharmacokinetics

We built the CUDA solver, wired Starling-Onsager to per-patient genomic variants, and optimised for multi-GPU throughput.

04

Validation & deployment

We ran the 340-patient retrospective study and deployed on a hybrid on-premise/cloud infrastructure with native PACS integration.

We had the data and we had the science. What we lacked was a team that could turn them into something that actually works in a clinical setting. Zeklar did exactly that, with an engineering rigour you rarely find outside academia.

Dr. M. Ferrante, Director of Translational Research, Confidential Consortium

// The impact

KAIROS aims to cut chemotherapy adverse events by ~30%, to be demonstrated through clinical validation with partner oncology centres. Today it is an in-silico research prototype, eligible for digital-health funding programmes; it is not a medical device.

NEXTVORTEX

ZEKLAR software technology 4.0/5.0

Let's talk
ZEKLAR · SOFTWARE THAT MAKES THE DIFFERENCE.
ZEKLAR · VAT L81419031T · info@zeklar.com© 2026