| 1. |
- Thambawita, V., et al.
(författare)
-
SinGAN-Seg: Synthetic training data generation for medical image segmentation
- 2022
-
Ingår i: PLoS ONE. - : Public Library of Science (PLoS). - 1932-6203. ; 17:5 May
-
Tidskriftsartikel (refereegranskat)abstract
- Analyzing medical data to find abnormalities is a time-consuming and costly task, particularly for rare abnormalities, requiring tremendous efforts from medical experts. Therefore, artificial intelligence has become a popular tool for the automatic processing of medical data, acting as a supportive tool for doctors. However, the machine learning models used to build these tools are highly dependent on the data used to train them. Large amounts of data can be difficult to obtain in medicine due to privacy reasons, expensive and time-consuming annotations, and a general lack of data samples for infrequent lesions. In this study, we present a novel synthetic data generation pipeline, called SinGAN-Seg, to produce synthetic medical images with corresponding masks using a single training image. Our method is different from the traditional generative adversarial networks (GANs) because our model needs only a single image and the corresponding ground truth to train. We also show that the synthetic data generation pipeline can be used to produce alternative artificial segmentation datasets with corresponding ground truth masks when real datasets are not allowed to share. The pipeline is evaluated using qualitative and quantitative comparisons between real data and synthetic data to show that the style transfer technique used in our pipeline significantly improves the quality of the generated data and our method is better than other state-of-the-art GANs to prepare synthetic images when the size of training datasets are limited. By training UNet++ using both real data and the synthetic data generated from the SinGAN-Seg pipeline, we show that the models trained on synthetic data have very close performances to those trained on real data when both datasets have a considerable amount of training data. In contrast, we show that synthetic data generated from the SinGAN-Seg pipeline improves the performance of segmentation models when training datasets do not have a considerable amount of data. All experiments were performed using an open dataset and the code is publicly available on GitHub.
|
|
| 2. |
- Calvani, P., et al.
(författare)
-
Infrared spectroscopy of two-dimensional electron systems
- 2019
-
Ingår i: European Physical Journal: Special Topics. - : Springer Science and Business Media LLC. - 1951-6401 .- 1951-6355. ; 228:3, s. 669-673
-
Tidskriftsartikel (refereegranskat)abstract
- © 2019, EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature. We have used grazing-angle infrared spectroscopy with polarized radiation to detect the Berreman effect (BE) in the two-dimensional electron systems (2DES) which form spontaneously at two interfaces: one between an amorphous film LaAlO3 and its SrTiO3 substrate (LAO/STO), and another at the interface between the topological insulator (TI) Bi2Se3 and its sapphire substrate. In both systems we have thus extracted the 2DES parameters at different temperatures. In the quasi-2DES under amorphous LAO, the surface density ns is higher than under crystalline LAO, while the mobility is nearly the same and the thickness d is 7 nm. In ultrapure Bi2Se3 on sapphire, preliminary data provided d
|
|
| 3. |
- Sinaei, Sima, et al.
(författare)
-
ELC-ECG : Efficient LSTM cell for ECG classification based on quantized architecture
- 2021
-
Ingår i: Proceedings - IEEE International Symposium on Circuits and Systems. - : Institute of Electrical and Electronics Engineers Inc.. - 9781728192017 ; May
-
Konferensbidrag (refereegranskat)abstract
- Long Short-Term Memory (LSTM) is one of the most popular and effective Recurrent Neural Network (RNN) models used for sequence learning in applications such as ECG signal classification. Complex LSTMs could hardly be deployed on resource-limited bio-medical wearable devices due to the huge amount of computations and memory requirements. Binary LSTMs are introduced to cope with this problem. However, naive binarization leads to significant accuracy loss in ECG classification. In this paper, we propose an efficient LSTM cell along with a novel hardware architecture for ECG classification. By deploying 5-level binarized inputs and just 1-level binarization for weights, output, and in-memory cell activations, the delay of one LSTM cell operation is reduced 50x with about 0.004% accuracy loss in comparison with full precision design of ECG classification.
|
|
