The accurate detection of leukocytes is the basis for the diagnosis of blood system diseases. However, diagnosing leukocyte disorders by doctors is time-consuming and requires extensive experience. Automated detection methods with high accuracy can improve detection efficiency and provide recommendations to inexperienced doctors. Current methods and instruments either fail to automate the identification process fully or have low performance and need suitable leukocyte data sets for further study. To improve the current status, we need to develop more intelligent strategies. This paper investigates fulfilling high-performance automatic detection for leukocytes using a deep learning-based method. We established a new dataset more suitable for leukocyte detection, containing 6273 images (8595 leukocytes) and considering nine common clinical interference factors. Based on the dataset, the performance evaluation of six mainstream detection models is carried out, and a more robust ensemble model is proposed. The mean of average precision (mAP) @IoU = 0.50:0.95 and mean of average recall (mAR)@IoU = 0.50:0.95 of the ensemble model on the test set are 0.853 and 0.922, respectively. The detection performance of poor-quality images is robust. For the first time, it is found that the ensemble model yields an accuracy of 98.84% for detecting incomplete leukocytes. In addition, we also compared the test results of different models and found multiple identical false detections of the models, then provided correct suggestions for the clinic.
Satisfactory biomarkers for screening and early diagnosis of lung cancer remain scarce and require further investigation. The aim of the present study was to examine the changes of the biochemical and protein composition in the serum and pleural effusion from lung cancer and lung infection (bacterial pneumonia) patients. A total of 92 patients with lung cancer, 38 with bacterial pneumonia and 42 healthy controls were enrolled in the study. The serum levels of cholesterol, apolipoprotein A and transthyretin (TTR) in the lung cancer patients were higher than that of the lung infection patients (P<0.05). The levels of TTR were higher, whereas the activity of adenosine deaminase (ADA) was lower in the pleural effusion from the lung cancer patients compared to the lung infection patients (P<0.05). Furthermore, the pleural effusion/serum TTR ratios in the lung cancer patients were higher, whereas the ratios of ADA were lower (P<0.05). By matrix-assisted laser desorption/ionization time-of-flight mass spectrometry analysis, four major peaks corresponding to native TTR, Sul-TTR, Cys-TTR and Cysgly-TTR were observed in the serum of the lung cancer and lung infection patients. A significant increase was found in the proportion of Cysgly-TTR in the pleural effusion from the patients with lung cancer. The data indicated that a combination of pleural effusion/serum TTR ratios and modified TTR may be beneficial for the differential diagnosis between lung cancer and lung infection.
The application of THz technique into detecting myocardial amyloidosis exists some challenges: the irregularity of tissues, the overlap of spectra, the few sampling points and low signal-to-noise ratio (SNR) of spectra. In this paper, we propose an efficient method to construct the THz dataset of myocardial amyloidosis, which includes 4319 samples. Besides, we propose a convolutional neural networks (CNNs) based framework to detect myocardial amyloidosis. Benefiting from both the THz technique and CNNs, an accuracy of 98.37%, a precision of 97.94%, a recall of 98.85% and a F1-score of 98.39% are achieved. In 10 independent repeated tests, the extreme deviations of the four indicators obtained are within 0.65%, which proves that the proposed model has prominent stability. Furthermore, we have compared it with the existing sensors, which shows that our method provides innovation and improvement in the simplicity of operation, the efficiency of detection and the reliability of discrimination. To verify the comparability of CNNs, we compare it with the state-of-the-art (SOTA) algorithms and has achieved all the best performance under different number of samples. Not only that, under the circumstances of fewest sampling points and lowest SNR, the CNNs can still achieve a F1-score of 93.74% and 94.68%, respectively. This preliminary work indicates that the combination of THz-TDS and CNNs is of great potential in accurately and efficiently detecting myocardial amyloidosis.
The clinical characteristics of excreted tumor cells can be found in the urine of bladder cancer patients, meaning the identification of tumor cells in urine can assist in bladder cancer diagnosis. The presence of white blood cells and epithelial cells in the urine interferes with the recognition of tumor cells. In this paper, a technique for detecting cancer cells in urine based on microfluidics provides a novel approach to bladder cancer diagnosis. The bladder cancer cell line (T24) and MeT-5A were used as positive bladder tumor cells and non-tumor cells, respectively. The practicality of the tumor cell detection system based on microfluidic cell chip detection technology is discussed. The tumor cell (T24) concentration was around 1 × 104 to 300 × 104 cells/mL. When phosphate buffer saline (PBS) was the diluted solution, the tumor cell detected rate was 63–71% and the detection of tumor cell number stability (coefficient of variation, CV%) was 6.7–4.1%, while when urine was the diluted solution, the tumor cell detected rate was 64–72% and the detection of tumor cell number stability (CV%) was 6.3–3.9%. In addition, both PBS and urine are tumor cell dilution fluid solutions. The sample was analyzed at a speed of 750 microns per hour. Based on the above experiments, a system for detecting bladder cancer cells in urine by microfluidic analysis chip technology was reported. The rate of recognizing bladder cancer cells reached 68.4%, and the speed reached 2 mL/h.
1 October 2021 in Computational Biology and Chemistry