• Journal of Embryo Transfer
• /
• v.11 no.3
• /
• pp.291-300
• /
• 1996

Real time B-mode ultrasound was used to detect the early conceptus in 187 Korean native cattles between days 10 and 60 after last insemination. The ultrasound diagnostic findings were systemically confirmed by palpation per rectum after the 60th day of last insemination. The embryonic vesicle and the embryo proper within the veside were first visible on mean day fl and 23, respectively. The heartbeat of the embryo proper could be detected on day 26, and the limb buds, placentomes, amnion, fetal movement, umbilical cord, optic area and split hooves were first visible on day 33, 34, 34, 44.5, 45, 32 and 48, respectively. The mean length of embryo proper was 3.8mm on day 23 which later increased to 56. 6rnrn on day 60. When ultrasound was used to detect the conceptus between days 20 and 30 after insemination and palpation per rectum after the 60th day of insemination, the accuracy rates of pregnancy detection by ultrasound scanning at days 20, 22, 24, 26, 28, 30 were 44.4, 69.2, 78.6, 87.5, 90.0, 93.3%. In summary, the early pregnancy diagnosis of Korean native cattle with ultrasound appears high accuracy rates. It is considered that ultrasound can be used in veterinary practice well.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2304-2320
• /
• 2021

Intelligently detecting anomalies in health sensor data streams (e.g., Electrocardiogram, ECG) can improve the development of E-health industry. The physiological signals of patients are collected through sensors. Timely diagnosis and treatment save medical resources, promote physical health, and reduce complications. However, it is difficult to automatically classify the ECG data, as the features of ECGs are difficult to extract. And the volume of labeled ECG data is limited, which affects the classification performance. In this paper, we propose a Generative Adversarial Network (GAN)-based deep learning framework (called CAB) for heart arrhythmia classification. CAB focuses on improving the detection accuracy based on a small number of labeled samples. It is trained based on the class-imbalance ECG data. Augmenting ECG data by a GAN model eliminates the impact of data scarcity. After data augmentation, CAB classifies the ECG data by using a Bidirectional Long Short Term Memory Recurrent Neural Network (Bi-LSTM). Experiment results show a better performance of CAB compared with state-of-the-art methods. The overall classification accuracy of CAB is 99.71%. The F1-scores of classifying Normal beats (N), Supraventricular ectopic beats (S), Ventricular ectopic beats (V), Fusion beats (F) and Unclassifiable beats (Q) heartbeats are 99.86%, 97.66%, 99.05%, 98.57% and 99.88%, respectively. Unclassifiable beats (Q) heartbeats are 99.86%, 97.66%, 99.05%, 98.57% and 99.88%, respectively.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.19 no.9
• /
• pp.1010-1019
• /
• 2008

In this paper, we present a noise analysis and measurement results of a bio-radar system that can detect human heartbeat and respiration signals. The noise analysis including various phase noise effects is very important in designing the bio-radar system, since the frequency difference between the received signal and local oscillator is very small and the received power is very low. All of the noise components in a bio-radar system are considered from the point of view of SNR. From this analysis, it can be concluded that the phase noise due to antenna leakage is a dominant factor and is a function of range correlation. Therefore, the phase noise component with range correlation effect, which is the most important noise contribution, is measured using the measurement setup and compared with the calculated results. From the measurement results, our measurement setup can measure a closed-in phase noise of a free-running oscillator. Based on these results, it is possible to design a 2.4 GHz bio-radar system quantitatively which has a detection range of 50 cm and low power of 1 mW without additional PLL circuits.