-
AI Imaging Lab: Development & Validation of Segmentation and Detection Models
Karolinska Institutet (KI)
## Overview
This service supports SMEs and researchers in developing, training, and validating AI models for medical imaging applications. Hosted at SMAILE, Karolinska Institutet, it covers segmentation, detection, and classification tasks using CT, MRI, nuclear images, multimodal images, ultrasound, histology, and microscopic images. The pipeline includes data curation, evaluation of labeling strategy, model selection and training, evaluation metric selection, and model performance benchmarking.
We provide expertise and support in:
- AI model training and optimization for medical imaging
- Validation using clinical datasets and standard metrics (using publicly available datasets, datasets available through data agreements, and internal datasets at KI, depending on the case)
- Clinical Relevance and Comparison with the state of the art in research and clinical practice
- Imaging biomarkers studies for diagnosis, prognosis, and prediction applications
### How can the service help you?
The service ensures your imaging AI model performs reliably and is aligned with clinical expectations. Whether you’re entering the pre-clinical testing phase or seeking validation to secure investment or regulatory approval, this service equips you with a rigorous evaluation and feedback report.
### How the service will be delivered?
Available both virtually and physically. Imaging data can be reviewed remotely through a secure data transfer process. On-site collaboration is also possible for sensitive datasets or model development, evaluation, and validation. A typical project takes 4–6 weeks.
---
## Additional information
### Provider description
SMAILE is the digital health core facility at Karolinska Institute. It offers interdisciplinary support in AI for medical imaging, data analytics, and system validation, partnering with leading institutions under the Swedish TEF-Health node.
### Technical description
The imaging model pipeline is extensively validated by using a curated benchmark set and standard evaluation frameworks such as well-established quantification metrics for object detection, image segmentation, and classification tasks.. Annotation quality is reviewed, and model performance is benchmarked against open or reference models. Standard medical image processing tools such as PyDicom, ANTs, and ITK, as well as community-driven open-sourced frameworks such as MONAI, are the core components of our designed pipelines.
### Service customization
The service can focus on either model development, evaluation, or both. Datasets can be anonymized and securely shared, or analysis can be conducted in a local sandboxed environment.
View service →
-
Anatomic pathology: human prepared slides, data, analysis, expertise
Centre Hospitalier Universitaire De Rennes (CHU RENNES)
Anatomic pathology slides: preparation +/- associated patient clinical data +/- analysis +/- software evaluation +/- medical expertise; cancer, lesion, sample, biopsy, tissue, organ; histology, diagnosis, genetic; microscope, scanner
View service →
-
Datasets for Diagnostic Evaluation and AI Performance in Medical Specialties
Univerzitna Nemocnica Martin (University Hospital Martin)
This service provides tailored datasets for evaluating AI performance in key medical specialties, including sleep disorder diagnostics, gastroenterology, digital pathology, and radiology. The datasets reflect real clinical scenarios, supporting the validation of AI algorithms and assisting organizations in achieving regulatory compliance and clinical reliability. Each dataset includes structured data and annotations designed for training, testing, and comparing AI models in various clinical contexts.
Keywords: AI dataset, clinical diagnostics, digital pathology, radiology, gastroenterology, sleep disorders, performance validation.
View service →
-
Non-human genome sequencing
Karolinska Institutet (KI)
The Unit: Clinical Genomics Stockholm (https://www.scilifelab.se/units/clinical-genomics-stockholm/)
This infrastructure is part of Science for Life Laboratory (SciLifeLab) which is an academic collaboration between Swedish universities (including Karolinska Institutet) and a national research infrastructure with a focus on life science.
Clinical Genomics Stockholm unit is a research infrastructure for large-scale, genomics-based analyses using next generation sequencing technologies. The unit assists in translational research projects, in the translation of genomics-based tools to routine clinical care and also aims to improve the capability for national microbial surveillance and for pandemic preparedness.
The Service: Non-human genome sequencing
Sequencing of bacterial or viral whole genomes with the possibility of sequence analysis and strain typing.
Sequencing of SARS-CoV-2 amplicons.
Sequencing of metagenome samples (with PCR-free protocol).
This service encompasses:
o Consultation (Project design, Target capture design)
o Sample management
o DNA& RNA sequencing (microbial genomes, metagenomics)
o Bioinformatics (Bioinformatic analysis on data generated at CG)
Equipment
Our unit has access to specialized equipment that enable us to factilitate the translation of new high-throughput techniques into clinical use. These include, but are not restricted to:
• Various automatic robotic systems (BRAVO NGS Workstation, Hamilton NGS Star)
• Various systems for QC, quantification and fragment analyses (TapeStation, Quantification using fluorescent assay (Qubit, Quantit))
• Instrumentation for real time PCR and ddPCR (BioRad qPCR)
• Instrumentation for Multispectral imaging (Saphyr optical mapper)
• Sequencing platforms:
o Illumina (NovaSeq 6000, NovaSeq 6000 Dx, NovaSeq X)
o Nanopore (PromethION “access via NGI”)
o PacBio (Revio)
View service →
-
Single Cell Multiomic profiling
Karolinska Institutet (KI)
The Unit: Eukaryotic Single Cell Genomics (https://www.scilifelab.se/units/eukaryotic-single-cell-genomics/)
The infrastructure is part of Science for Life Laboratory (SciLifeLab) which is an academic collaboration between Swedish universities (including Karolinska Institutet) and a national research infrastructure with a focus on life science.
Single-cell genomics technologies are rapidly advancing and have proven to give new insights into cell type discovery and in the characterization of heterogeneity in tumors as well as in normal tissue.
The Eukaryotic Single Cell Genomics (ESCG) unit aims at providing high-throughput single cell transcriptomics services through a streamlined and complete single-cell RNA sequencing service. The user provide us with single cells, we process the samples, sequence and deliver annotated gene expression data.
• Study heterogeneity within putatively homogeneous cell populations
• Unbiased discovery of cell types in complex tissues
• Characterizing the cellular and genetic composition of tumors
The service: Single Cell Multiomic profiling
10X Genomics (droplet-based) single cell RNA sequencing in combination with single cell immune profiling (VDJ), AND/OR in combination with cell surface protein detection, OR in combination with ATAC (Assay for Transposase Accessible Chromatin) to analyze thousands of unique open chromatin fragments genome-wide with single cell resolution. ESCG provides access to technology, end-to-end support from help with project planning, quality check of your sample, cDNA library preparation to analysis and associated bioinformatics support.
View service →
-
single cell RNA sequencing on CRISPR modified cells
Karolinska Institutet (KI)
The Units: CRISPR Functional Genomics (CFG) (https://www.scilifelab.se/units/crispr-functional-genomics/) and Eukaryotic Single Cell Genomics (ESCG) (https://www.scilifelab.se/units/eukaryotic-single-cell-genomics/)
Both infrastructures are part of Science for Life Laboratory (SciLifeLab) which is an academic collaboration between Swedish universities (including Karolinska Institutet) and a national research infrastructure with a focus on life science.
Both units, CFG and ESCG, have designed a pipeline of CRISPR pooled screens together with single cell RNASeq readout (Perturb-Seq, CROP-Seq). This allows to assess gene expression phenotypes at the single cell level following inactivation of a pool of genes, using Cas-9 expressing cells.
Service span the entire pooled screening process, from Cas-cell line generation and library virus creation, via phenotypic selection by FACS or live/dead, to single cell library preparation, NGS and data analysis.
View service →
