An AI tool for the analysis of medical images based on neural networks, developed by Laife Reply

Artificial intelligence is revolutionising our world, with the healthcare sector being no exception. Within this context, X-RAIS is an AI tool for the analysis of medical images based on neural networks, developed by Laife Reply. X-RAIS supports doctors in the medical record compilation phase, automatically calling attention to suspicious areas and the related classification, with the aim of reducing the number of incorrect diagnoses and improving the efficiency of the end-to-end diagnostic process. Currently X-RAIS specialises in reading mammograms and is able to determine the degree of breast density, highlight any lesions that may be present and describe the relative morphological characteristics.

  • strip-0


    X-RAIS is a Deep Learning platform verticalised on different diagnostic methods (e.g. radiography, ultrasound, MRI) and on specific anatomical districts.
    X-RAIS supports medical diagnosis processes through the application of Medical Image Recognition techniques, identifying with high accuracy areas of the image on which the doctor should focus his or her attention.


    According to a study conducted by the Breast Cancer Surveillance Consortium, when it comes to mammography screening, only 84.4% of malignant tumours are actually diagnosed at the first examination.
    It is therefore essential to strengthen and enhance the medical diagnosis processes that rely on images, by providing a decision support tool to help doctors (without replacing them) in the medical record compilation process, reducing the number of incorrect diagnoses, decreasing the time it takes to complete the diagnosis and therefore increasing the production capacity.

  • strip-2


    X-RAIS has been designed to work with DICOM (Digital Imaging and Communications in Medicine) images, without in any way altering the original image; X-R can therefore be integrated into various PACS (Picture Archiving and Communication System) and becomes part of the radiologist's normal working processes.
    With each new recording request on PACS, the results of the analysis carried out by X-RAIS are shown automatically. Specifically, a map is superimposed over the DICOM image in which the suspicious areas are highlighted (localisation) and, for every anomaly, a description characterising the lesion is presented (classification) according to the ACR BI-RADS classification. Moreover, thanks to the determination of the level of density and the identification of positive areas in the breast, X-RAIS is able to effectively organise the operational workflow in the report, with particular benefits in the screening processes.


    The Artificial Intelligence and cognitive services offered by cloud architectures allow X-RAIS to be applied effectively in any context. X-RAIS ensures data confidentiality and integrity, essential elements in light of the new privacy regulations that have come into force (GDPR). X-RAIS can be used as a service, based on Pay-Per-Use or annual licensing models and in specific cases, the special architecture that X‑RAIS benefits from also supports an on-prem installation or use.

  • strip-4


    Laife Reply combines the extensive expertise of the Reply Group in the innovative technologies sector (Big Data, Cloud Computing, Digital Media and the Internet of Things) and a broad know-how in the clinical/medical areas to provide Artificial Intelligence platforms specifically designed to meet the needs of the Healthcare sector, with the aim of strengthening and enhancing medical diagnosis processes, providing decision support in clinical areas and helping to improve patients’ life-styles.