Quibim

In this research for JMIR Publications, our team tested 11 AutoML frameworks, 6 general-purpose and 5 radiomics-specific, across 10 public and private CT/MRI datasets with different anatomies, sizes, and endpoints, all under the same standardized cross-validation.   Most radiomics-specific frameworks were excluded from the final performance comparison: obsolete or computationally impractical. General-purpose frameworks were clearly easier to implement. Simplatab, a radiomics-specific tool with a no-code interface, achieved the best balance of performance and efficiency (mean test AUC 78.46%, runtime 1.1h), but wasn't statistically better than the most computationally intensive general-purpose options.   No single framework clearly emerged. AutoML for radiomics still needs tools built for its real challenges: survival analysis, feature harmonization, and reproducibility across the full workflow. https://bit.ly/4qd5OXx #ArtificialIntelligence #MedicalImaging #Radiomics #MachineLearning

Manual interpretation of prostate MRI scans is time-consuming, and every additional step between image acquisition and intervention takes time you simply don’t have.   DynaCAD powered by QP-Prostate® is now available to bridge that gap.   Fully automated AI-driven analysis—segmentation, lesion detection, and risk classification—integrated directly into the DynaCAD workflow, so that the process from image acquisition to intervention simply flows.   We’ll be at the FTS & SURS in Arlington, VA, 17-19 September. Book a demo in advance: https://lnkd.in/evPryKCX Nathan Clark | Joshua Moss | Edward Ramirez R.T. (R)(MR) B.S. | Eric Draper | Michael S. | Christopher Miller | Tom Tynes | Paavo Immonen | Kevin Coady | CJ Carlson | Ángel Alberich-Bayarri | Alberto Valiente Montoro #ProstateCancer #RadiologyAI #Urology #MedicalImaging

From France to Uruguay, leading radiologists are already using QP-Prostate® in their daily practice to detect prostate cancer faster and with greater confidence.   It does not replace the radiologist, it provides an additional layer of safety through an automated second reading, reduces inter-observer variability, and helps detect lesions even in the most difficult-to-evaluate areas. Because at the end of every reading is a patient — and that's exactly why precision and speed matter. Philippe PUECH | Margarita García Fontes | Kai Vilanova MD, PhD #ProstateCancer #MedicalImaging #AI

Can AI assistance improve radiologists' detection of clinically significant prostate cancer? According to new data on QP-Prostate® published in the European Journal of Radiology, it can.   Four radiologists (with between 1 and 25 years' experience) interpreted MRI scans of 52 men who subsequently underwent a prostatectomy, which served as ground truth. Each case was interpreted in three ways: radiologists alone, AI alone, and radiologists with AI assistance. The combination of AI + radiologists delivered the best results: ▪️ The AUC at lesion level improved for all four readers with AI assistance (from +0.013 to +0.037, all confidence intervals exclude zero) ▪️ Sensitivity increased significantly in three of the four readers, particularly in the least experienced reader (from 0.41 to 0.49) ▪️ Specificity didn't change, meaning no increase in unnecessary biopsies Read the full paper: https://bit.ly/4x84oQl #ProstateCancer #ArtificialIntelligence #AIinHealthcare

Breast MRI has become an essential tool for women with dense breast tissue, high genetic risk, or a strong family history, offering higher sensitivity for breast cancer lesion detection and characterization.   Interpretation remains complex, with high variability between readers, and as a result, up to 4 out of 5 biopsied lesions are not cancer, leading to unnecessary procedures, increased healthcare costs, and anxiety for the patient.   QP-Breast® automatically detects, segments, and characterizes suspicious lesions, helping radiologists interpret breast MRI with more confidence and consistency, all within a clear report integrated into the PACS workflow.   Want to see it in action? https://bit.ly/4fI3eUT #BreastCancer #BreastMRI #HealthcareInnovation #WomensHealth

We're happy to share that our CEO and Co-Founder, Dr. Ángel Alberich-Bayarri has been invited to speak at the 2026 Society of Robotic Surgery | SRS Annual Meeting, taking place July 23–26 in Hollywood, FL. A huge thank you to the SRS organization, and especially to Dr. Victoria Bird, for this invitation and for recognizing the growing role of AI in advancing precision surgery. "𝘘𝘶𝘪𝘣𝘪𝘮 𝘈𝘐: 𝘗𝘳𝘰𝘴𝘵𝘢𝘵𝘦 𝘤𝘢𝘯𝘤𝘦𝘳 𝘪𝘥𝘦𝘯𝘵𝘪𝘧𝘪𝘤𝘢𝘵𝘪𝘰𝘯 𝘸𝘪𝘵𝘩 𝘈𝘐 𝘢𝘭𝘨𝘰𝘳𝘪𝘵𝘩𝘮𝘴" (𝘍𝘰𝘤𝘢𝘭 𝘛𝘩𝘦𝘳𝘢𝘱𝘺 𝘵𝘳𝘢𝘤𝘬) 📅 Sunday, July 26 🕚 11:40–11:50 AM EST 📍 Room 303 If you're a urologist attending #SRS2026, join us to see how Quibim's AI algorithms help in the whole prostate cancer patient pathway, bringing evidence-based precision to focal therapy decisions. See you there! #Urology #ProstateCancer #ArtificialIntelligence

More than 3.4 billion people suffer from a neurological condition, but not everyone has access to the same standard of brain health care.   The most advanced imaging techniques are helping to bridge that gap: the same scan, lasting just 14 minutes, now provides far more information than before. A recent article published in Nature shows how this translates into practice. “Multiplexed MRI” (MRx) extracts up to 22 structural, physiological and molecular biomarkers from a single clinical scan, characterizing the state of tissues with greater precision than conventional magnetic resonance imaging. In patients with brain tumors and multiple sclerosis, the authors report an improvement in the classification of lesions and tumors, laying the foundations for truly quantitative imaging techniques geared towards precision medicine.   The incorporation of these tools into daily clinical practice is what transforms a scientific breakthrough into better brain health for more people: more information with the same effort, as the earlier a disease is detected, the wider the range of treatment options available.   https://lnkd.in/eqdSUEUs   #BrainHealthDay #MedicalImaging #AIinHealth

Focal therapy for prostate cancer has been on the rise for over a decade, but the long-term oncological outcomes remained unclear. A new study published in European Urology begins to fill that gap.   3,477 patients, 14 centres in the UK, the HEAT and ICE registries, treated with HIFU or cryotherapy between 2004 and 2024. At 10 years: prostate cancer-specific mortality of 0.13%; metastasis rate of 3.3%. The vast majority of the cohort (96%) had intermediate- or high-risk disease — a marked contrast with most focal therapy studies, which typically exclude higher-risk patients.   The authors highlight what is still missing: a validated risk calculator for patient selection and validated MRI scoring systems to detect local recurrence earlier. Both rely on precise localization of the lesion on MRI, which is where QP-Prostate comes in: lesion detection aligned with PI-RADS v2.1 and structured reports for prostate MRI. https://lnkd.in/enTJvX-N #ProstateCancer #Urology #MedicalImaging #DigitalHealth #FocalTherapy

Consulting firms with no medical device credentials are already implementing artificial intelligence tools in healthcare centres across Europe, not only for administrative tasks, but also to suggest diagnoses and influence clinical decisions. The moment a system influences diagnosis or treatment, it becomes a medical device, and in Europe this falls within the scope of the Medical Devices Regulation (MDR): a risk-based framework that requires clinical evidence, traceability and post-market surveillance. So we need to ask ourselves some simple questions. Who is technically responsible for this system? Who has authorised its clinical deployment? Who oversees it once it is in place? And if a patient suffers any harm, who is actually liable? If no one can provide a clear answer, the tool is not ready for clinical use. 🗞️ Read the full article here: https://lnkd.in/eS_DCxXg The European Business Review #AIinHealthcare #HealthTech #DigitalHealth

According to 23-year follow-up data from the European Prostate Cancer Screening Study (NEJM Group), prostate cancer screening can reduce mortality by up to 13%. Dr. Jeronimo Barrera Ortega (Hospital Universitario Nuestra Señora del Rosario) walks through what this means for clinical practice and where imaging fits into the pathway: from identifying who should undergo screening to supporting radiologists when a suspicious finding appears on an MRI scan. Watch the full conversation: https://lnkd.in/ewV__2WP #ProstateCancer #MedicalImaging #AI #Radiology