Transfer Learning-Based Multi-Scale Denoising Convolutional Neural Network for Prostate Cancer Detection.

BACKGROUND: Prostate cancer is the 4th most common type of cancer. To reduce the workload of medical personnel in the medical diagnosis of prostate cancer and increase the diagnostic accuracy in noisy images, a deep learning model is desired for prostate cancer detection. METHODS: A multi-scale denoising convolutional neural network (MSDCNN) model was designed for prostate cancer detection (PCD) that is capable of noise suppression in images. The model was further optimized by transfer learning, which contributes domain knowledge from the same domain (prostate cancer data) but heterogeneous datasets. Particularly, Gaussian noise was introduced in the source datasets before knowledge transfer to the target dataset. RESULTS: Four benchmark datasets were chosen as representative prostate cancer datasets. Ablation study and performance comparison between the proposed work and existing works were performed. Our model improved the accuracy by more than 10% compared with the existing works. Ablation studies also showed average improvements in accuracy using denoising, multi-scale scheme, and transfer learning, by 2.80%, 3.30%, and 3.13%, respectively. CONCLUSIONS: The performance evaluation and comparison of the proposed model confirm the importance and benefits of image noise suppression and transfer of knowledge from heterogeneous datasets of the same domain.

A [(M2)6L8] metal-organic polyhedron with high CO2 uptake and efficient chemical conversion of CO2 under ambient conditions.

A new metal-organic polyhedron with a high surface area of 407 m2 g-1, possessing high CO2 uptake, is reported, which is synthesized using 4-connected Cu2(CO2)4 paddle-wheel moieties and 3-connected semi-rigid tripodal carboxylates. This material possesses a high density of Cu(II) Lewis acidic sites and demonstrates excellent performance as a heterogeneous catalyst for the chemical fixation of CO2 into cyclic carbonates under ambient conditions.

The applications and prospects of hydrophobic metal-organic frameworks in catalysis.

In recent years, large numbers of hydrophobic/superhydrophobic metal-organic frameworks (MOFs) have been developed. These hydrophobic MOFs not only retain rich structural variety, highly crystalline frameworks, and uniform micropores, but they also have lower affinity towards water and boosted hydrolytic stability. Until now, there were two main strategies to prepare hydrophobic MOFs, including a one-step method and post-synthesis modification (PSM). PSM was an often-used strategy for preparing hydrophobic MOFs. Hydrophobic MOFs showed unique advantages when used as catalysts for various categories of reactions. Herein, recent research advances relating to hydrophobic MOFs in the catalytic field are presented. The catalytic activities of hydrophobic MOFs and corresponding hydrophilic ones are also compared, and the superiority of hydrophobic MOFs or MOF materials as catalysts in 10 reactions is discussed. Finally, the advantages of hydrophobic MOFs as catalysts or auxiliary materials are summarized and promising future developments of hydrophobic MOFs are highlighted.

A Precise Drunk Driving Detection Using Weighted Kernel Based on Electrocardiogram.

Globally, 1.2 million people die and 50 million people are injured annually due to traffic accidents. These traffic accidents cost $500 billion dollars. Drunk drivers are found in 40% of the traffic crashes. Existing drunk driving detection (DDD) systems do not provide accurate detection and pre-warning concurrently. Electrocardiogram (ECG) is a proven biosignal that accurately and simultaneously reflects human's biological status. In this letter, a classifier for DDD based on ECG is investigated in an attempt to reduce traffic accidents caused by drunk drivers. At this point, it appears that there is no known research or literature found on ECG classifier for DDD. To identify drunk syndromes, the ECG signals from drunk drivers are studied and analyzed. As such, a precise ECG-based DDD (ECG-DDD) using a weighted kernel is developed. From the measurements, 10 key features of ECG signals were identified. To incorporate the important features, the feature vectors are weighted in the customization of kernel functions. Four commonly adopted kernel functions are studied. Results reveal that weighted feature vectors improve the accuracy by 11% compared to the computation using the prime kernel. Evaluation shows that ECG-DDD improved the accuracy by 8% to 18% compared to prevailing methods.

A high fuel consumption efficiency management scheme for PHEVs using an adaptive genetic algorithm.

A high fuel efficiency management scheme for plug-in hybrid electric vehicles (PHEVs) has been developed. In order to achieve fuel consumption reduction, an adaptive genetic algorithm scheme has been designed to adaptively manage the energy resource usage. The objective function of the genetic algorithm is implemented by designing a fuzzy logic controller which closely monitors and resembles the driving conditions and environment of PHEVs, thus trading off between petrol versus electricity for optimal driving efficiency. Comparison between calculated results and publicized data shows that the achieved efficiency of the fuzzified genetic algorithm is better by 10% than existing schemes. The developed scheme, if fully adopted, would help reduce over 600 tons of CO2 emissions worldwide every day.

High accuracy localization of long term evolution based on a new multiple carrier noise model.

A high accuracy localization technique using Long Term Evolution (LTE) based on a new and accurate multiple carrier noise model has been developed. In the noise consideration, the LTE multiple carriers phase noise has been incorporated so that a new and accurate noise model is achieved. An experiment was performed to characterize the phase noise of carriers at 2 GHz. The developed noise model was incorporated into LTE localization analysis in a high traffic area in Hong Kong to evaluate the accuracy of localization. The evaluation and analysis reveals that the new localization method achieves an improvement of about 10% accuracy comparing to existing widely adopted schemes.