Skin Lesion Classification on Imbalanced Data Using Deep Learning with Soft Attention.

Today, the rapid development of industrial zones leads to an increased incidence of skin diseases because of polluted air. According to a report by the American Cancer Society, it is estimated that in 2022 there will be about 100,000 people suffering from skin cancer and more than 7600 of these people will not survive. In the context that doctors at provincial hospitals and health facilities are overloaded, doctors at lower levels lack experience, and having a tool to support doctors in the process of diagnosing skin diseases quickly and accurately is essential. Along with the strong development of artificial intelligence technologies, many solutions to support the diagnosis of skin diseases have been researched and developed. In this paper, a combination of one Deep Learning model (DenseNet, InceptionNet, ResNet, etc) with Soft-Attention, which unsupervisedly extract a heat map of main skin lesions. Furthermore, personal information including age and gender are also used. It is worth noting that a new loss function that takes into account the data imbalance is also proposed. Experimental results on data set HAM10000 show that using InceptionResNetV2 with Soft-Attention and the new loss function gives 90 percent accuracy, mean of precision, F1-score, recall, and AUC of 0.81, 0.81, 0.82, and 0.99, respectively. Besides, using MobileNetV3Large combined with Soft-Attention and the new loss function, even though the number of parameters is 11 times less and the number of hidden layers is 4 times less, it achieves an accuracy of 0.86 and 30 times faster diagnosis than InceptionResNetV2.

Development of the typical driving cycle for buses in Hanoi, Vietnam.

This paper develops a typical driving cycle for buses in Hanoi that does not require the deconstruction of the natural driving patterns. Real velocity-time data were collected along 15 routes in the inner city. The raw velocity-time series were preprocessed to remove errors, and smooth and denoise the data. These data, then, were tested for stationary behavior before being used in the construction of the driving cycle based on Markov chain theory. The 14 representative parameters of the driving cycle, including vehicle-specific power, which were extracted from 33 driving cycle parameters using the hierarchical agglomerative clustering method, were used to integrate the features of realistic driving patterns into the typical driving cycle. The conformity of the developed driving cycle with the real-world driving data was evaluated by the speed-acceleration frequency distribution (SAFD). A typical driving cycle for buses in Hanoi with a SAFD of 13.2% was developed. This is the first driving cycle developed for buses in Vietnam. Implications: A typical driving cycle was developed for the first time for buses in Hanoi. With the deviation in speed-acceleration frequency distribution (SAFD) reaching to 13.2%, the developed driving cycle reflects well the overall real-world driving data in the city. This driving cycle, therefore, can be applied for the development of the country-specific emission factors and emission inventories for buses which are a very good tool as well as useful information for integrated air quality management in Hanoi.

Development of the specific emission factors for buses in Hanoi, Vietnam.

This paper develops the specific emission factors for buses in the real-world traffic conditions in the inner city of Hanoi, Vietnam. An engine stationary cycle consisting of 14 modes was developed based on the typical driving cycle of Hanoi buses which had been constructed with the application of Markov chain theory. This is the first engine stationary emissions test cycle constructed for heavy-duty engine in Vietnam. Based on this cycle, the country-specific emission factors (CSEFs) of air pollutants including CO, HC, NOx, CO2, and PM for buses in Hanoi have been developed using the emission measurements on the engine test bed. It is found that almost all developed emission factors are higher than those derived from the emission measurements to the ECE R49 on the same engine. These emission factors, therefore, can be used to improve the quality of the emission inventory of buses in Hanoi.