A Deep Learning Approach for Surface Crack Classification and Segmentation in Unmanned Aerial Vehicle Assisted Infrastructure Inspections.

Surface crack detection is an integral part of infrastructure health surveys. This work presents a transformative shift towards rapid and reliable data collection capabilities, dramatically reducing the time spent on inspecting infrastructures. Two unmanned aerial vehicles (UAVs) were deployed, enabling the capturing of images simultaneously for efficient coverage of the structure. The suggested drone hardware is especially suitable for the inspection of infrastructure with confined spaces that UAVs with a broader footprint are incapable of accessing due to a lack of safe access or positioning data. The collected image data were analyzed using a binary classification convolutional neural network (CNN), effectively filtering out images containing cracks. A comparison of state-of-the-art CNN architectures against a novel CNN layout "CrackClassCNN" was investigated to obtain the optimal layout for classification. A Segment Anything Model (SAM) was employed to segment defect areas, and its performance was benchmarked against manually annotated images. The suggested "CrackClassCNN" achieved an accuracy rate of 95.02%, and the SAM segmentation process yielded a mean Intersection over Union (IoU) score of 0.778 and an F1 score of 0.735. It was concluded that the selected UAV platform, the communication network, and the suggested processing techniques were highly effective in surface crack detection.

Reinvestigation of the Two-step Synthesis of Sevoflurane.

Improvements in the two-step synthesis of 1,1,1,3,3,3-hexafluoro-2- (fluoromehoxy)propane (Sevoflurane) that result in the product cost reduction, safety level enhancement and positive environmental impacts are described. This process consists of chloromethylation reaction of 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) followed by a halogen-exchange fluorination. This is the first synthesis of Sevoflurane in Iran which was successfully scaled up. During this work, several improvements have been achieved by optimization of the reaction time, the amount of consumed starting materials and solvents and work up procedure while keeping the yield and purity intact. The reaction time of the first step (24 h) was diminished to 4 h. (19)F NMR spectroscopy was used to investigate the rate of the reaction in the first step and to evaluate the influence of different parameters mentioned on the achieved improvements.

Prediction of sulphide build-up in filled sewer pipes.

Millions of dollars are being spent worldwide on the repair and maintenance of sewer networks and wastewater treatment plants. The production and emission of hydrogen sulphide has been identified as a major cause of corrosion and odour problems in sewer networks. Accurate prediction of sulphide build-up in a sewer system helps engineers and asset managers to appropriately formulate strategies for optimal sewer management and reliability analysis. This paper presents a novel methodology to model and predict the sulphide build-up for steady state condition in filled sewer pipes. The proposed model is developed using a novel data-driven technique called evolutionary polynomial regression (EPR) and it involves the most effective parameters in the sulphide build-up problem. EPR is a hybrid technique, combining genetic algorithm and least square. It is shown that the proposed model can provide a better prediction for the sulphide build-up as compared with conventional models.

Infection Pattern of Neutropenic Patients in Post-chemotherapy Phase of Acute Leukemia Treatment.

Neutropenia following chemotherapy regimens in leukemia patients is of major concern since it makes these patients vulnerable to infections. If we can identify which germs are causing these infections, they can be annihilated or, at least, the most appropriate antibiotic therapy can be started immediately, even before we have the results of the culture. This retrospective multi-center study took place in 2012 and included patients with acute leukemia who had already undergone chemotherapy and who had been febrile for at least 16 hours. In order to assess the type of infection, different environments were chosen and the results were compared by t-test and χ(2) tests. This study took place in four hospitals in Tehran and Ahwaz, Iran. The study population was made up of 89 patients: 37 with acute lymphoblastic leukemia and 52 with acute myeloid leukemia. The results revealed that blood was the most common site of infection. From all our positive cultures, it was seen that 85.4% of them had gram-negative bacteria with a dominance of E. coli of 25.8% over the other colonies. Also, antibiograms revealed the sensitivity of almost all the gram-negatives to amino glycosides. In contrast with most of the literature, in our patients, gram-negatives are the most common cause of infection and, therefore, administering amino glycosides would be the safest antibiotic therapy to prescribe before culture results are available.