Present-day federated learning (FL) systems deployed over edge networks have to consistently deal with a large number of workers with high degrees of heterogeneity in data and/or computing capabilities. This diverse set of workers necessitates the development of FL algorithms that allow: (1) flexible worker participation that grants the workers' capability to engage in training at will, (2) varying number of local updates (based on computational resources) at each worker along with asynchronous communication with the server, and (3) heterogeneous data across workers. To address these challenges, in this work, we propose a new paradigm in FL called "Anarchic Federated Learning" (AFL). In stark contrast to conventional FL models, each worker in AFL has complete freedom to choose i) when to participate in FL, and ii) the number of local steps to perform in each round based on its current situation (e.g., battery level, communication channels, privacy concerns). However, AFL also introduces significant challenges in algorithmic design because the server needs to handle the chaotic worker behaviors. Toward this end, we propose two Anarchic FedAvg-like algorithms with two-sided learning rates for both cross-device and cross-silo settings, which are named AFedAvg-TSLR-CD and AFedAvg-TSLR-CS, respectively. For general worker information arrival processes, we show that both algorithms retain the highly desirable linear speedup effect in the new AFL paradigm. Moreover, we show that our AFedAvg-TSLR algorithmic framework can be viewed as a meta-algorithm for AFL in the sense that they can utilize advanced FL algorithms as worker-and/or server-side optimizers to achieve enhanced performance under AFL. We validate the proposed algorithms with extensive experiments on real-world datasets.

Machine learning relies on the availability of a vast amount of data for training. However, in reality, most data are scattered across different organizations and cannot be easily integrated under many legal and practical constraints. In this paper, we introduce a new technique and framework, known as federated transfer learning (FTL), to improve statistical models under a data federation. The federation allows knowledge to be shared without compromising user privacy, and enables complimentary knowledge to be transferred in the network. As a result, a target-domain party can build more flexible and powerful models by leveraging rich labels from a source-domain party. A secure transfer cross validation approach is also proposed to guard the FTL performance under the federation. The framework requires minimal modifications to the existing model structure and provides the same level of accuracy as the nonprivacy-preserving approach. This framework is very flexible and can be effectively adapted to various secure multi-party machine learning tasks.

This paper defines a simple protocol for competitive and quantified evaluation of electromagnetic tracking systems such as the NDI Aurora (A) and Ascension microBIRD with dipole transmitter (B). It establishes new methods and a new phantom design which assesses the reproducibility and allows comparability with different tracking systems in a consistent environment. A machined base plate was designed and manufactured in which a 50 mm grid of holes was precisely drilled for position measurements. In the center a circle of 32 equispaced holes enables the accurate measurement of rotation. The sensors can be clamped in a small mount which fits into pairs of grid holes on the base plate. Relative positional/orientational errors are found by subtracting the known distances/ rotations between the machined locations from the differences of the mean observed positions/ rotation. To measure the influence of metallic objects we inserted rods made of steel (SST 303, SST 416), aluminum, and bronze into the sensitive volume between sensor and emitter. We calculated the fiducial registration error and fiducial location error with a standard stylus calibration for both tracking systems and assessed two different methods of stylus calibration. The positional jitter amounted to 0.14 mm(A) and 0.08 mm(B). A relative positional error of 0.96 mm +/- 0.68 mm, range -0.06 mm; 2.23 mm(A) and 1.14 mm +/- 0.78 mm, range -3.72 mm; 1.57 mm(B) for a given distance of 50 mm was found. The relative rotation error was found to be 0.51 degrees (A)/0.04 degrees (B). The most relevant distortion caused by metallic objects results from SST 416. The maximum error 4.2 mm(A)/ > or = 100 mm(B) occurs when the rod is close to the sensor(20 mm). While (B) is more sensitive with respect to metallic objects, (A) is less accurate concerning orientation measurements. (B) showed a systematic error when distances are calculated.

No abstract

To improve image quality and CT number accuracy of fast-scan low-dose cone-beam computed tomography (CBCT) through a deep-learning convolutional neural network (CNN) methodology for head-and-neck (HN) radiotherapy. Fifty-five paired CBCT and CT images from HN patients were retrospectively analysed. Among them, 15 patients underwent adaptive replanning during treatment, thus had same-day CT/CBCT pairs. The remaining 40 patients (post-operative) had paired planning CT and 1st fraction CBCT images with minimal anatomic changes. A 2D U-Net architecture with 27-layers in 5 depths was built for the CNN. CNN training was performed using data from 40 post-operative HN patients with 2080 paired CT/CBCT slices. Validation and test datasets include 5 same-day datasets with 260 slice pairs and 10 same-day datasets with 520 slice pairs, respectively. To examine the impact of differences in training dataset selection and network performance as a function of training data size, additional networks were trained using 30, 40 and 50 datasets. Image quality of enhanced CBCT images were quantitatively compared against the CT image using mean absolute error (MAE) of Hounsfield units (HU), signal-to-noise ratio (SNR) and structural similarity (SSIM). Enhanced CBCT images reduced artifact distortion and improved soft tissue contrast. Networks trained with 40 datasets had imaging performance comparable to those trained with 50 datasets and outperformed those trained with 30 datasets. Comparison of CBCT and enhanced CBCT images demonstrated improvement in average MAE from 172.73 to 49.28 HU, SNR from 8.27 to 14.25 dB, and SSIM from 0.42 to 0.85. The image processing time is 2 s per patient using a NVIDIA GeForce GTX 1080 Ti GPU. The proposed deep-leaning methodology was fast and effective for image quality enhancement of fast-scan low-dose CBCT. This method has potential to support fast online-adaptive re-planning for HN cancer patients.