Rapid fiber-detection technique by artificial intelligence in phase-contrast microscope images of simulated atmospheric samples.

Since the manufacture, import, and use of asbestos products have been completely abolished in Japan, the main cause of asbestos emissions into the atmosphere is the demolition and removal of buildings built with asbestos-containing materials. To detect and correct asbestos emissions from inappropriate demolition and removal operations at an early stage, a rapid method to measure atmospheric asbestos fibers is required. The current rapid measurement method is a combination of short-term atmospheric sampling and phase-contrast microscopy counting. However, visual counting takes a considerable amount of time and is not sufficiently fast. Using artificial intelligence (AI) to analyze microscope images to detect fibers may greatly reduce the time required for counting. Therefore, in this study, we investigated the use of AI image analysis for detecting fibers in phase-contrast microscope images. A series of simulated atmospheric samples prepared from standard samples of amosite and chrysotile were observed using a phase-contrast microscope. Images were captured, and training datasets were created from the counting results of expert analysts. We adopted 2 types of AI models-an instance segmentation model, namely the mask region-based convolutional neural network (Mask R-CNN), and a semantic segmentation model, namely the multi-level aggregation network (MA-Net)-that were trained to detect asbestos fibers. The accuracy of fiber detection achieved with the Mask R-CNN model was 57% for recall and 46% for precision, whereas the accuracy achieved with the MA-Net model was 95% for recall and 91% for precision. Therefore, satisfactory results were obtained with the MA-Net model. The time required for fiber detection was less than 1 s per image in both AI models, which was faster than the time required for counting by an expert analyst.

Development of rapid and highly accurate method to measure concentration of fibers in atmosphere using artificial intelligence and scanning electron microscopy.

AIM: We aimed to develop a measurement method that can count fibers rapidly by scanning electron microscopy equipped with an artificial intelligence image recognition system (AI-SEM), detecting thin fibers which cannot be observed by a conventional phase contrast microscopy (PCM) method. METHODS: We created a simulation sampling filter of airborne fibers using water-filtered chrysotile (white asbestos). A total of 108 images was taken of the samples at a 5 kV accelerating voltage with 10 000X magnification scanning electron microscopy (SEM). Each of three expert analysts counted 108 images and created a model answer for fibers. We trained the artificial intelligence (AI) using 25 of the 108 images. After the training, the AI counted fibers in 108 images again. RESULTS: There was a 12.1% difference between the AI counting results and the model answer. At 10 000X magnification, AI-SEM can detect 87.9% of fibers with a diameter of 0.06-3 µm, which is similar to a skilled analyst. Fibers with a diameter of 0.2 µm or less cannot be confirmed by phase-contrast microscopy (PCM). When observing the same area in 300 images with 1500X magnification SEM-as listed in the Asbestos Monitoring Manual (Ministry of the Environment)-with 10 000X SEM, the expected analysis time required for the trained AI is 5 h, whereas the expected time required for observation by an analyst is 251 h. CONCLUSION: The AI-SEM can count thin fibers with higher accuracy and more quickly than conventional methods by PCM and SEM.