| 1. |
- Yu, Wenjin, et al.
(författare)
-
Deep Learning-Based Classification of Cancer Cell in Leptomeningeal Metastasis on Cytomorphologic Features of Cerebrospinal Fluid
- 2022
-
Ingår i: Frontiers in Oncology. - : Frontiers Media SA. - 2234-943X. ; 12, s. 1-11
-
Tidskriftsartikel (refereegranskat)abstract
- Background: It is a critical challenge to diagnose leptomeningeal metastasis (LM), given its technical difficulty and the lack of typical symptoms. The existing gold standard of diagnosing LM is to use positive cerebrospinal fluid (CSF) cytology, which consumes significantly more time to classify cells under a microscope.Objective: This study aims to establish a deep learning model to classify cancer cells in CSF, thus facilitating doctors to achieve an accurate and fast diagnosis of LM in an early stage.Method: The cerebrospinal fluid laboratory of Xijing Hospital provides 53,255 cells from 90 LM patients in the research. We used two deep convolutional neural networks (CNN) models to classify cells in the CSF. A five-way cell classification model (CNN1) consists of lymphocytes, monocytes, neutrophils, erythrocytes, and cancer cells. A four-way cancer cell classification model (CNN2) consists of lung cancer cells, gastric cancer cells, breast cancer cells, and pancreatic cancer cells. Here, the CNN models were constructed by Resnet-inception-V2. We evaluated the performance of the proposed models on two external datasets and compared them with the results from 42 doctors of various levels of experience in the human-machine tests. Furthermore, we develop a computer-aided diagnosis (CAD) software to generate cytology diagnosis reports in the research rapidly.Results: With respect to the validation set, the mean average precision (mAP) of CNN1 is over 95% and that of CNN2 is close to 80%. Hence, the proposed deep learning model effectively classifies cells in CSF to facilitate the screening of cancer cells. In the human-machine tests, the accuracy of CNN1 is similar to the results from experts, with higher accuracy than doctors in other levels. Moreover, the overall accuracy of CNN2 is 10% higher than that of experts, with a time consumption of only one-third of that consumed by an expert. Using the CAD software saves 90% working time of cytologists.Conclusion: A deep learning method has been developed to assist the LM diagnosis with high accuracy and low time consumption effectively. Thanks to labeled data and step-by-step training, our proposed method can successfully classify cancer cells in the CSF to assist LM diagnosis early. In addition, this unique research can predict cancer’s primary source of LM, which relies on cytomorphologic features without immunohistochemistry. Our results show that deep learning can be widely used in medical images to classify cerebrospinal fluid cells. For complex cancer classification tasks, the accuracy of the proposed method is significantly higher than that of specialist doctors, and its performance is better than that of junior doctors and interns. The application of CNNs and CAD software may ultimately aid in expediting the diagnosis and overcoming the shortage of experienced cytologists, thereby facilitating earlier treatment and improving the prognosis of LM.
|
|
| 2. |
- Huang, Ze Hua, et al.
(författare)
-
Utilizing Haematococcus pluvialis to simulate animal meat color in high-moisture meat analogues: Texture quality and color stability
- 2024
-
Ingår i: Food Research International. - 0963-9969 .- 1873-7145. ; 175
-
Tidskriftsartikel (refereegranskat)abstract
- The effect of Haematococcus pluvialis (HP) (0.25∼1.25 %) as a colorant during high moisture extrusion (50 %) on the texture and microstructural properties of soy protein-based high moisture meat analogs (HMMA) was evaluated. Furthermore, the stability of HP-induced meat like color of the HMMA as a function of light exposure, freeze/thawing, frozen storage and cooking temperature and duration was investigated. The addition of HP reduced the elasticity of HMMA but enhanced its hardness, chewiness, and resilience. HP addition at low levels promoted the flexible and disordered regions within the protein secondary structure while excessive HP addition was unfavorable for protein cross-linking. The optimal degree of texturization was achieved with 0.75 % HP. Sensory evaluations revealed that HMMA with 1 %HP had a color similar to fresh beef sirloin, while HMMA with 0.25 % HP had a color closer to fresh pork loin. Light exposure induced the greatest color loss of the meat analogs compared with the cooking and frozen storage. The a* value of HMMA containing 1.25 % HP decreased by 30 % during the 14 days of light exposure. Frozen storage at darkness efficiently preserved the meat-like color of the extrudates. Overall, HP was found as promising colorant for HMMA production but the storage condition of the extrudates should be carefully optimized.
|
|
| 3. |
- Tan, Zhifei, et al.
(författare)
-
Numerical study of the aggregate contact effect on the complex modulus of asphalt concrete
- 2022
-
Ingår i: Materials & design. - : Elsevier BV. - 0264-1275 .- 1873-4197. ; 213
-
Tidskriftsartikel (refereegranskat)abstract
- Asphalt concrete (AC) is a composite material consisting of binder, aggregates and air voids. The quantitative effect of aggregate-to-aggregate contact on the mechanical performance of AC is an important and complex issue, which has not been fully understood yet. To fill this gap, this study aims to characterize the aggregate contacts in AC and evaluate their effects on the viscoelastic behavior of AC through micromechanical finite element (FE) modeling. To this end, 3D microstructural models were generated through digital image processing (DIP) method and aggregate contacts were captured in the model via contact zone (CZ) elements. A CZ model was proposed and verified by a parametric study to identify the viscoelastic properties of CZ elements, while the viscoelastic properties of matrix phase were determined through laboratory tests. Steady-state dynamic (SSD) analysis was then conducted to investigate the macro-scale viscoelastic response of AC. It was found that the proposed modeling approach captures the measured response accurately. Accounting for aggregate contacts results in higher predicted AC dynamic moduli and lower phase angles, thus improving the agreement between modeling and experimental results. The numerical model developed in this study provides a promising approach for investigating the effect of aggregate contacts on the mechanical performance of AC.
|
|
