About the Customer
Starting as a single medical centre in 1987, Aster DM Healthcare has metamorphosized into becoming the leading healthcare provider across the Middle East and Africa (MENA) region. With over 27 hospitals, 115 clinics and 223 pharmacies in 7 countries, Aster is touted as one of the fastest growinghealthcare conglomerates offering an expansive portfolio of performance driven healthcare services. Headquartered in Dubai, the hospital network currently encompasses approximately 19,000 employees and has built a reputation for constantly being on a look out for innovative opportunities, setting new yardsticks in healthcare advancement.
The Pain Point
In 2021, Aster DM Healthcare announced the creation of a separate digital healthcare vertical and appointed Brandon Rowberry, a global leader in corporate innovation and strategy, as its CEO. Creation of a dedicated digital health division aligns with the hospital’s long-term plan of building digital transformation and innovation as central pillars to achieve enhanced health outcomes. From integrating robust hospital information systems to applying the Da Vinci Robotic Surgical System for executing minimally invasive surgery with 3D vision, Aster has been at the forefront of incorporating cutting-edge digital solutions across the healthcare continuum.
As one would expect, the organisation has been very agile and is actively undertaking digital healthcare research initiatives in collaboration with prominent players worldwide. Both doctors and data scientists across the Aster network have demonstrated keen interest towards undertaking research projects that test applications of Artificial Intelligence (AI) and Federated Learning (FL) models across the healthcare value chain. Interacting with multiple teams at Aster, we were confident that CARPL could most definitely serve as a gateway that allows the organisation to provide both medical and digital expertise to global AI initiatives focused on building future-ready healthcare systems.
Given CARPL’s infrastructure agnostic deployment mechanism, it was deployed on Aster compute infrastructure and integrated with various PACS systems across the organisation. This enabled the following projects:
Using CARPL to provide radiology expertise to a leading AI software company
- Aster leveraged the CARPL platform to participate in a multi-reader, multi-site retrospective, study with a leading global AI medical software company.
- The objective of the clinical trial was to examine feasibility of the company’s investigational device that detects pulmonary nodules in chest CT scans.
- About 10 leading doctors from Aster participated in the study to provide radiology expertise in the form of both preliminary readings and final ground truth reads concerning presence of pulmonary nodule on the CTs.
- By the end of both phase 1 and phase 2 of the trial which lasted approximately six months, the radiologists had executed ~2,000 reads on the CARPL platform.
Leveraging CARPL’s data storage mechanisms and technical expertise to create a cloud-based data lake
- Aster’s visionary team of data scientists used CARPL as a middle layer to curate and transfer data from PACS at multiple hospital sites.
- Post curation, teams from both CARPL and Aster worked in tandem to create an application layer that can push data from PACS to Aster’s newly developed cloud-based data lake.
- Aster has been actively working on building federated learning models on the data lake, the data for which is curated and pulled using CARPL.
While software companies globally are steadily developing AI / ML solutions for the healthcare industry, incorporating clinical nuances at every stage of development is crucial to create a product that is both compliant with regulatory standards and is deployment ready. CARPL aims to address this concern by serving as a singular platform that connects both doctors and data scientists to work together seamlessly on multicentric clinical trials of cutting-edge medical software products. CARPL’s various modules – data management, search, annotation, AI integration, testing and monitoring – make it flexible enough to be used across healthcare providers for a wide array of use-cases. In Aster’s case, by providing a secure and robust environment for systematic flow of medical data from multiple sites, CARPL’s vision is to serve as the digital backbone as the organisation builds in-house cloud-based repositories for enhanced R&D and clinical outcomes.