Bonescreen offers a novel, AI-based technology for non-invasive extraction of biomarkers from existing medical images to assist radiologists and clinicians in evaluating bone health.
clinicians in diagnosing*
SpineR facilitates your spine processing*
SpineQ assists clinicians in diagnosing*
Bonescreen spineQ assists clinicians in diagnosing osteoporosis from almost any existing routine CT containing the spine. spineQ provides a comprehensive spine report, including a level-wise annotation, a calibrated, reproducible bone mineral density measurement and osteoporotic fracture identification. It is designed to shorten reporting times and improve report quality by highlighting abnormal findings relevant for osteoporosis that are frequently overlooked.
SpineR facilitates your spine processing*
Bonescreen spineR is easily integrated in many environments as the backbone of spine processing. In CT scans, SpineR segments and labels all vertebrae, identifies subregions and special points of interest, like commonly needed in automated surgery planning, PACS-overlays, finite element modelling, 3D printing and more. Bonescreen spineR outperforms current state of the art methods for spine segmentation in a computationally efficient way without the need for a GPU. Bonescreen readily identifies transitional vertebrae and all enumeration anomalies.
Bonescreen fits your needs
Optimised for your working environment
Empowering radiologists with precise, quantitative insights into bone diseases, such as osteoporosis, for more informed diagnoses.
Seamlessly integrating context information and anatomical guidance, enhancing the efficiency and completeness of patient records.
Enabling bone researchers with advanced tools for precise segmentation and landmark detection, accelerating their research efforts.
Providing clinicians with a quick, at-a-glance way to identify patients at risk, enhancing patient care and decision-making.
Intraoperative provision of level specific quantitative information on bone quality to surgeons in real time.
Bonescreen is backed by peer-reviewed research
August 23, 2022
August 31, 2021
June 20, 2023
Financial & scientific partners
Bonescreen was founded in 2022 and is headquartered in Munich.
Collectively, we have 25 years of experience in radiology and 15 of experience in software engineering and machine learning.
Why do we need
The doubling of the population older than 60 years during the next three decades will bring an explosion of bone-health-related conditions. Among these, Osteoporosis represents the most common bone disease in humans.
Osteoporosis is a disease that can be treated effectively, if diagnosed early. However, effective screening programs are currently missing. This is what Bonescreen aims to improve with our AI-based opportunistic screening software.
In a similar way, we are working on a solution to detect malignant bone lesions much earlier and more accurate compared to the current standard. All these patients may highly benefit from an earlier and more accurate diagnosis.
Preventive healthcare remains an underserved market
On average, European countries invest circa 3-4% of their total healthcare spend in preventive measures. The remaining 97% of costs goes to treating diseases.
This is too low, claims the OECD. Estimates imply that healthcare costs could be reduced by 30% if more preventive action were taken. In light of the €55bn the EU spends every year on the treatment of osteoporotic fractures, this implies cost savings of €15bn.
Let alone the prevented loss of quality in life of the more than 4 million European patients who suffer an osteoporotic fracture each year. At Bonescreen, we build prevention. An impact tool to provide millions of Europeans access to preventive measures.
is to prevent fractures and make dignified ageing possible: Bonescreen will increase screening rates for osteoporosis by utilizing existing medical imaging data. With this, Bonescreen contributes to a prevention-based healthcare system away from primarily treating complications [Löffler MT et al., Eur Radiol. 2021; Sollmann N et al., J Bone Miner Res. 2022].
of women and 20% of men older than 50 years will have an osteoporotic fracture during their lifetime. In Europe, osteoporotic fractures are the fourth leading cause of morbidity, after ischemic heart disease, dementia and lung cancer, however before chronic obstructive pulmonary disease and ischemic stroke.
of patients who should be treated according to current osteoporosis guidelines do NOT receive their appropriate treatment in Germany, with almost similar numbers for whole Europe. This huge treatment gap is mainly caused by a diagnostic gap, which in turn is due to a lack of appropriate screening strategies [McCloskey E et al., Osteoporos Int 2021].
were spent in Europe to treat osteoporosis-related fragility fractures in 2019. Only €1.7bn were spent on medication that can prevent subsequent fractures. These numbers don’t account for additional personal and societal costs due to pain, reduced mobility, disability, lost working days and increased mortality [Kanis A et al., Arch Osteoporos. 2021].
specificity was observed for opportunistic quantitative CT to predict vertebral fractures during follow-up in a study by Löffler et al.. The overall diagnostic performance, measured as “area under the curve” significantly outperformed dual x-ray absorptiometry in multiple studies [Chalhoub D et al., Bone 2016;Löffler MT et al., Eur Radiol. 2019].
despite widely available and effective treatment options, more than 97% of osteoporosis-related health care costs are spent on treating fractures and only less than 3% on prevention. As osteoporosis is well treatable in early stages, many medical societies advocate a population-based screening program based on bone density measurements.
of humans have anomalies at the spine, where additional lumbar vertebrae are the most frequent ones among Europeans, while thoracolumbar transitional vertebrae dominate in Asia. Our fully automated labelling and segmentation readily and reliably identifies those anomalies and is consistent with functional nerve anatomy [Ishiguro K et al. Sci Rep 2020].