Search

10th IEEE International Conference on Healthcare Informatics (IEEE ICHI) -- JUN 11-14, 2022 -- Rochester, MN, Pretrained language models elevated with in-domain corpora show impressive results in biomedicine and clinical NLP tasks in English. However, there is minimal work in low-resource languages. This work introduces the BioBERTurk family, three pretrained models in Turkish for biomedicine. To evaluate models, we also introduce a labeled dataset to classify radiology reports of CT exams. Our first model was initialized from BERTurk and pretrained with biomedical corpus. The second model again continues to pretrain the general BERT model with a corpus of Ph.D. theses on radiology to test the effect of the task-related text. The final model combines radiology and biomedicine corpora with the corpus of BERTurk and pretrained a BERT model from scratch. F-scores of our models in the radiology resort classification are 92.99, 92.75, and 89.49 respectively. As far as we know, this is the first model that evaluates the effect of small size in-domain corpus in pretraining from scratch., IEEE,Mayo Clin, Dept AI & Informat,Mayo Clin, Robert D & Patricia E Kern Ctr Sci Hlth Care Delivery,Mayo Clin Platform,NSF,IEEE Comp Soc Tech Comm Intelligent Informat,Journal Healthcare Informat Res,Hlth Data Sci, Tensorflow Research Cloud (TRC), We would like to acknowledge the support we received from the Tensorflow Research Cloud (TRC) team 4 in providing access to TPUv3 units.

In this study, the design of an omnidirectional and circularly polarized waveguide antenna used as a transceiver in telemetry/telecommand applications of satellite systems is described. Instead of separate transmitter (TX) and receiver (RX) antennas used in space segment, the proposed antenna can realize transmit and receive operations (TX/RX) simultaneously, which results in the reduction in the number of antennas. The proposed antenna has two sets of inclined slots on circular waveguides with different radii. The below slots operate at TX frequencies around 11.75 GHz, and above slots make an effective radiation at RX frequencies around 14 GHz. In the antenna design, while metallic disks extending away from circular waveguides improve axial ratio performance, a special dual-band rectangular-to-circular waveguide mode converter provides better omnidirectionality. The designed antenna has 10 dB impedance bandwidth more than 150 MHz and 415 MHz at TX and RX frequencies. In azimuth plane, gain variation is lower than 2 dBi, axial ratio is also lower than 1.63 dB at both frequency bands. In elevation plane, minimum gain is found as -5 dBi and axial ratio is below almost 3 dB within 55° beamwidth.

Transport mode detection is a subset of activity recognition. It is important for urban transportation planning and finding CO 2 footprint for individuals. However, daily usage of these applications are limited due to the power consumption of mobile devices. By the wide spread usage of wearable sensors, it will be crucial to develop systems that have smaller model size and less computational complexity. This paper presents a novel system that exploits n-grams which are widely used in natural language processing for extracting features of mobile sensor signals. This is the first study at the literature that provides smaller and faster activity recognition models by using n-grams. Our results demonstrate that, feature extraction with n-grams is less complex and creates 4 times smaller models even though its accuracy is comparable with classical methods. In addition, if n-gram based features and some of traditional features are combined, accuracy of the model achieves 89% which outperforms the research that use shallow machine learning approaches. Although this study focused on transport mode detection, we believe that n-gram based features could be used for several signal classification tasks.

There is a need for more skilled people in cybersecurity, and initiatives are being taken all over the world to improve the education of both students and professionals. This includes establishing cyber ranges and using them for training. Despite several cyber range projects going on, there is still a need for platforms that are easily accessible for both teachers and students. This paper presents ∗DefAtt∗, which is a highly automatized platform for red team and blue team games, particularly focusing on network-based attacks and their defenses. The platform supports a high degree of gamification, while at the same time giving the participants realistic feedback through the emulation of a Security Operations Center (SOC). The platform is open-source and available for everyone to use and contribute to.

Handwriting recognition is a common problem in applications such as converting handwritten documents to digital documents and authentication. Existing approaches utilize the handwritten document as the most important clue in recognition process. In this work secondary observation is used instead of the written document. Secondary observation becomes the only source of information when handwritten document cannot be observed directly or additional clues are needed for authentication. Secondary observation refers to the observation of the movement of the pen or the hand during writing. Three deep learning models are proposed to predict what is written by observing only the pen movement. The proposed solutions are the first for this problem. The result of the experiments presents a comparison of the success of three proposed models and shows that secondary observation is useful to solve the problem. Among the proposed models, the best results are obtained by using LSTM architecture which is frequently used to learn from temporal data.

Payload capacity, range and flight duration of a Unmanned Aerial System (UAS) depends mainly on thrust system. Thrust system of a typical electrical multi-rotor UAS consists of electric motors, rotors and electric speed controllers (ESCs). Electric motors are easy to use and their response time is short, which is highly desired and required for control, but they limit potential payload capacity, range and flight duration. In this paper, we propose a Quad-Jet VTOL (vertical take-off and land) UAS with a thrust system consisting of four mini jet engines. Such a vehicle could be useful heavy-lift applications. The UAS model, controllers, thrust direction system (to overcome the response time disadvantage and to improve motion) are presented, along with a first prototype system.

In this work, a W-band SiGe BiCMOS Low Noise Amplifier (LNA) that achieves 18 dB peak gain at 88 GHz with 6 mW of quiescent power consumption is presented. The LNA has a measured 3-dB bandwidth of 23 GHz and the minimum noise figure is 4.9 dB at 93 GHz and lower than 6.2 dB across the W-band. The LNA’s gain can be raised to 22dB by increasing the supply voltages with a negligible variance in matching performance at the cost of increased DC power consumption. Inductors were preferred to transmission lines for a reduction in the chip area, down to an effective area of 0.2mm 2 . Coherency between simulations and measurement results show that the LNA is suitable for use in a state-of-the-art low-power radiometer system.

This study presents a novel design of a rectangular patch antenna with rhombic shaped slots (RSS) on the radiating part and five additional slots on the ground plane to create defected ground structure. The first prototype is a conventional inset fed rectangular patch antenna operating at 3.51 GHz which is among the early phase 5G frequency spectrum. The antenna has a return loss level of -37.97 dB and a 10 dB bandwidth of 114.8 MHz. In order to achieve lower return loss levels and improved bandwidth RSS are etched on the radiating part and ground plane. Detailed analysis of the effects of RSS on the characteristics of patch antenna are performed. The results have shown that the final antenna has a 10 dB bandwidth of 116.4 MHz with a return loss level of -55.42 dB which is a 1600 Hz increase in bandwidth and 17.45 dB decrease in return loss levels compared to the reference antenna. As a result of these improved radiation characteristics, the proposed antenna is a suitable candidate for the early phase 5G applications.