RESUMO
Self-driving cars and autonomous vehicles are revolutionizing the automotive sector, shaping the future of mobility altogether. Although the integration of novel technologies such as Artificial Intelligence (AI) and Cloud/Edge computing provides golden opportunities to improve autonomous driving applications, there is the need to modernize accordingly the whole prototyping and deployment cycle of AI components. This paper proposes a novel framework for developing so-called AI Inference Engines for autonomous driving applications based on deep learning modules, where training tasks are deployed elastically over both Cloud and Edge resources, with the purpose of reducing the required network bandwidth, as well as mitigating privacy issues. Based on our proposed data driven V-Model, we introduce a simple yet elegant solution for the AI components development cycle, where prototyping takes place in the cloud according to the Software-in-the-Loop (SiL) paradigm, while deployment and evaluation on the target ECUs (Electronic Control Units) is performed as Hardware-in-the-Loop (HiL) testing. The effectiveness of the proposed framework is demonstrated using two real-world use-cases of AI inference engines for autonomous vehicles, that is environment perception and most probable path prediction.
RESUMO
OBJECTIVE: The creation and exchange of patients' Electronic Healthcare Records have developed significantly in the last decade. Patients' records are however distributed in data silos across multiple healthcare facilities, posing technical and clinical challenges that may endanger patients' safety. Current healthcare sharing systems ensure interoperability of patients' records across facilities, but they have limits in presenting doctors with the clinical context of the data in the records. We design and implement a platform for managing provenance tracking of Electronic Healthcare Records based on blockchain technology, compliant with the latest healthcare standards and following the patient-informed consent preferences. METHODS: The platform leverages two pillars: the use of international standards such as Integrating the Healthcare Enterprise (IHE), Health Level Seven International (HL7) and Fast Healthcare Interoperability Resources (FHIR) to achieve interoperability, and the use of a provenance creation process that by-design, avoids personal data storage within the blockchain. The platform consists of: (1) a smart contract implemented within the Hyperledger Fabric blockchain that manages provenance according to the W3C PROV for medical document in standardised formats (e.g. a CDA document, a FHIR resource, a DICOM study, etc.); (2) a Java Proxy that intercepts all the document submissions and retrievals for which provenance shall be evaluated; (3) a service used to retrieve the PROV document. RESULTS: We integrated our decentralised platform with the SpiritEHR engine, an enterprise-grade healthcare system, and we stored and retrieved the available documents in the Mandel's sample CDA repository,1 which contained no protected health information. Using a cloud-based blockchain solution, we observed that the overhead added to the typical processing time of reading and writing medical data is in the order of milliseconds. Moreover, the integration of the Proxy at the level of exchanged messages in EHR systems allows transparent usage of provenance data in multiple health computing domains such as decision making, data reconciliation, and patient consent auditing. CONCLUSIONS: By using international healthcare standards and a cloud-based blockchain deployment, we delivered a solution that can manage provenance of patients' records via transparent integration within the routine operations on healthcare data.
