Lightweight Multi-Class Support Vector Machine-Based Medical Diagnosis System with Privacy Preservation.

Machine learning, powered by cloud servers, has found application in medical diagnosis, enhancing the capabilities of smart healthcare services. Research literature demonstrates that the support vector machine (SVM) consistently demonstrates remarkable accuracy in medical diagnosis. Nonetheless, safeguarding patients' health data privacy and preserving the intellectual property of diagnosis models is of paramount importance. This concern arises from the common practice of outsourcing these models to third-party cloud servers that may not be entirely trustworthy. Few studies in the literature have delved into addressing these issues within SVM-based diagnosis systems. These studies, however, typically demand substantial communication and computational resources and may fail to conceal classification results and protect model intellectual property. This paper aims to tackle these limitations within a multi-class SVM medical diagnosis system. To achieve this, we have introduced modifications to an inner product encryption cryptosystem and incorporated it into our medical diagnosis framework. Notably, our cryptosystem proves to be more efficient than the Paillier and multi-party computation cryptography methods employed in previous research. Although we focus on a medical application in this paper, our approach can also be used for other applications that need the evaluation of machine learning models in a privacy-preserving way such as electricity theft detection in the smart grid, electric vehicle charging coordination, and vehicular social networks. To assess the performance and security of our approach, we conducted comprehensive analyses and experiments. Our findings demonstrate that our proposed method successfully fulfills our security and privacy objectives while maintaining high classification accuracy and minimizing communication and computational overhead.

Evolving Robust Policy Coverage Sets in Multi-Objective Markov Decision Processes Through Intrinsically Motivated Self-Play.

Many real-world decision-making problems involve multiple conflicting objectives that can not be optimized simultaneously without a compromise. Such problems are known as multi-objective Markov decision processes and they constitute a significant challenge for conventional single-objective reinforcement learning methods, especially when an optimal compromise cannot be determined beforehand. Multi-objective reinforcement learning methods address this challenge by finding an optimal coverage set of non-dominated policies that can satisfy any user's preference in solving the problem. However, this is achieved with costs of computational complexity, time consumption, and lack of adaptability to non-stationary environment dynamics. In order to address these limitations, there is a need for adaptive methods that can solve the problem in an online and robust manner. In this paper, we propose a novel developmental method that utilizes the adversarial self-play between an intrinsically motivated preference exploration component, and a policy coverage set optimization component that robustly evolves a convex coverage set of policies to solve the problem using preferences proposed by the former component. We show experimentally the effectiveness of the proposed method in comparison to state-of-the-art multi-objective reinforcement learning methods in stationary and non-stationary environments.

Leukoencephalopathy with brainstem and spinal cord involvement and lactate elevation (LBSL): assessment of the involved white matter tracts by MRI.

BACKGROUND AND PURPOSE: Leukoencephalopathy with brain stem and spinal cord involvement and lactate elevation (LBSL) is a recently identified autosomal recessive disorder with early onset of symptoms and slowly progressive pyramidal, cerebellar and dorsal column dysfunction. LBSL is characterized by distinct white matter abnormalities and selective involvement of brainstem and spinal cord tracts. The purpose of this study is to assess the imaging features of the involved white matter tracts in cases of LBSL by MRI. PATIENTS AND METHODS: We retrospectively reviewed the imaging features of the selectively involved white matter tracts in sixteen genetically proven cases of leukoencephalopathy with brainstem and spinal cord involvement and elevated brain lactate (LBSL). All patients presented with slowly progressive cerebellar sensory ataxia with spasticity and dorsal column dysfunction. MRI of the brain and spine using 1.5 T machine and proton magnetic resonance spectroscopy (1H MRS) on the abnormal white matter were done to all patients. The MRI and MRS data sets were analyzed according to lesion location, extent, distribution and signal pattern as well as metabolite values and ratios in MRS. Laboratory examinations ruled out classic leukodystrophies. RESULTS: In all cases, MRI showed high signal intensity in T2-weighted and FLAIR images within the cerebral subcortical, periventricular and deep white matter, posterior limbs of internal capsules, centrum semiovale, medulla oblongata, intraparenchymal trajectory of trigeminal nerves and deep cerebellar white matter. In the spine, the signal intensity of the dorsal column and lateral cortico-spinal tracts were altered in all patients. The subcortical U fibers, globi pallidi, thalami, midbrain and transverse pontine fibers were spared in all cases. In 11 cases (68.8%), the signal changes were inhomogeneous and confluent whereas in 5 patients (31.2%), the signal abnormalities were spotty. MRI also showed variable signal abnormalities in the sensory and pyramidal tracts in addition to the brainstem and cerebellar connections. Proton MRS showed consistent elevation of the lactate within the abnormal white matter. CONCLUSION: Distinct MRI findings in the form of selective affection of subcortical and deep white matter tracts of the brain (involving the posterior limb of internal capsules and sparing the subcortical U fibers), dorsal column and lateral cortico-spinal tracts of the spinal cord should lead to the diagnosis of LBSL supported by the presence of lactate peak in 1H MRS. The disease can be confirmed by the analysis of the disease gene DARS2.