DMGM: deformable-mechanism based cervical cancer staging via MRI multi-sequence.

Objective.This study aims to leverage a deep learning approach, specifically a deformable convolutional layer, for staging cervical cancer using multi-sequence MRI images. This is in response to the challenges doctors face in simultaneously identifying multiple sequences, a task that computer-aided diagnosis systems can potentially improve due to their vast information storage capabilities.Approach.To address the challenge of limited sample sizes, we introduce a sequence enhancement strategy to diversify samples and mitigate overfitting. We propose a novel deformable ConvLSTM module that integrates a deformable mechanism with ConvLSTM, enabling the model to adapt to data with varying structures. Furthermore, we introduce the deformable multi-sequence guidance model (DMGM) as an auxiliary diagnostic tool for cervical cancer staging.Main results.Through extensive testing, including comparative and ablation studies, we validate the effectiveness of the deformable ConvLSTM module and the DMGM. Our findings highlight the model's ability to adapt to the deformation mechanism and address the challenges in cervical cancer tumor staging, thereby overcoming the overfitting issue and ensuring the synchronization of asynchronous scan sequences. The research also utilized the multi-modal data from BraTS 2019 as an external test dataset to validate the effectiveness of the proposed methodology presented in this study.Significance.The DMGM represents the first deep learning model to analyze multiple MRI sequences for cervical cancer, demonstrating strong generalization capabilities and effective staging in small dataset scenarios. This has significant implications for both deep learning applications and medical diagnostics. The source code will be made available subsequently.

Integer particle swarm optimization based task scheduling for device-edge-cloud cooperative computing to improve SLA satisfaction.

Task scheduling helps to improve the resource efficiency and the user satisfaction for Device-Edge-Cloud Cooperative Computing (DE3C), by properly mapping requested tasks to hybrid device-edge-cloud resources. In this paper, we focused on the task scheduling problem for optimizing the Service-Level Agreement (SLA) satisfaction and the resource efficiency in DE3C environments. Existing works only focused on one or two of three sub-problems (offloading decision, task assignment and task ordering), leading to a sub-optimal solution. To address this issue, we first formulated the problem as a binary nonlinear programming, and proposed an integer particle swarm optimization method (IPSO) to solve the problem in a reasonable time. With integer coding of task assignment to computing cores, our proposed method exploited IPSO to jointly solve the problems of offloading decision and task assignment, and integrated earliest deadline first scheme into the IPSO to solve the task ordering problem for each core. Extensive experimental results showed that our method achieved upto 953% and 964% better performance than that of several classical and state-of-the-art task scheduling methods in SLA satisfaction and resource efficiency, respectively.

A three-stage heuristic task scheduling for optimizing the service level agreement satisfaction in device-edge-cloud cooperative computing.

Device-edge-cloud cooperative computing is increasingly popular as it can effectively address the problem of the resource scarcity of user devices. It is one of the most challenging issues to improve the resource efficiency by task scheduling in such computing environments. Existing works used limited resources of devices and edge servers in preference, which can lead to not full use of the abundance of cloud resources. This article studies the task scheduling problem to optimize the service level agreement satisfaction in terms of the number of tasks whose hard-deadlines are met for device-edge-cloud cooperative computing. This article first formulates the problem into a binary nonlinear programming, and then proposes a heuristic scheduling method with three stages to solve the problem in polynomial time. The first stage is trying to fully exploit the abundant cloud resources, by pre-scheduling user tasks in the resource priority order of clouds, edge servers, and local devices. In the second stage, the proposed heuristic method reschedules some tasks from edges to devices, to provide more available shared edge resources for other tasks cannot be completed locally, and schedules these tasks to edge servers. At the last stage, our method reschedules as many tasks as possible from clouds to edges or devices, to improve the resource cost. Experiment results show that our method has up to 59% better performance in service level agreement satisfaction without decreasing the resource efficiency, compared with eight of classical methods and state-of-the-art methods.