Spectral intelligent detection for aflatoxin B1 via contrastive learning based on Siamese network.

Aflatoxins, harmful substances found in peanuts, corn, and their derivatives, pose significant health risks. Addressing this, the presented research introduces an innovative MSGhostDNN model, merging contrastive learning with multi-scale convolutional networks for precise aflatoxin detection. The method significantly enhances feature discrimination, achieving an impressive 97.87% detection accuracy with a pre-trained model. By applying Grad-CAM, it further refines the model to identify key wavelengths, particularly 416 nm, and focuses on 40 key wavelengths for optimal performance with 97.46% accuracy. The study also incorporates a task dimensionality reduction approach for continuous learning, allowing effective ongoing aflatoxin spectrum monitoring in peanuts and corn. This approach not only boosts aflatoxin detection efficiency but also sets a precedent for rapid online detection of similar toxins, offering a promising solution to mitigate the health risks associated with aflatoxin exposure.

An Investigation of 2D Spine Magnetic Resonance Imaging (MRI) with Compressed Sensing (CS).

OBJECTIVE: To investigate the feasibility of compressed sensing MRI (CS-MRI) in the application of 2D spinal imaging and compare its performance with conventional MR imaging (non-CS-MRI). METHODS: The CS imaging protocol was optimized on 5 volunteers. Non-CS-MRI and CS-MRI of 2D sagittal T1 weighted imaging (WI), Sag T2WI, and axial T2WI were performed for 71 patients (22 cervical, 8 thoracic, 41 lumbar MRI). Paired t tests were conducted to compare the total scan time. Three radiologists assessed image quality and lesion diagnosis independently. A Kendall W test was performed to assess interobserver agreement of the image quality scores and lesion diagnosis between readers. A nonparametric test (Wilcoxon test) was performed to compare the image quality. For lesion diagnosis, the interobserver and interstudy agreements were evaluated by kappa analysis. Paired t tests were conducted for SNR and CNR comparison. RESULTS: The mean scan time for spine CS-MRI (4 min 28.7 s ± 34.6 s) was significantly shorter than that with non-CS-MRI (7 min 21.3 s ± 38.7 s, t = - 47.464, P < 0.0001). CS-MRI achieved higher SNR and CNR than Non-CS-MRI in image quality assessment. Interobserver agreements of lesion diagnosis were excellent between non-CS-MRI and CS-MRI (kappa value from 0.913 to 1.000, P < 0.001). Interstudy agreements of lesion assessments were also excellent (kappa value = 1.000, with P < 0.001). CONCLUSION: CS-MRI spine imaging can significantly reduce the scan time, while maintaining comparable imaging quality to non-CS-MRI.

MRI evaluation of pulmonary lesions and lung tissue changes induced by tuberculosis.

OBJECTIVE: To evaluate the utility of magnetic resonance imaging (MRI) with an advanced motion correction technique in characterizing lung tissue changes and lesions induced by pulmonary tuberculosis (TB). METHODS: Sixty-three subjects with computed tomography (CT) features of pulmonary TB underwent lung MRI. All subjects with pulmonary TB were confirmed by acid-fast bacillus (AFB) testing or the detection of Mycobacterium tuberculosis. T2-weighted turbo spin echo (TSE) sequence MRI with the MultiVane motion correction technique was used to image the lungs. Routine lung CT images were obtained as reference. MRI and CT images were reviewed by multiple readers independently. The performance of MRI in depicting abnormalities induced by pulmonary TB and their morphological changes were evaluated and compared with the performance of CT. RESULTS: Lung MRI found pulmonary abnormalities in all 63 TB subjects, with satisfactory quality. With the implementation of MultiVane for T2-weighted TSE sequences to reduce the motion correction effect, MRI showed excellent agreement with CT in detecting abnormal imaging features of pulmonary TB (κ=0.88, p<0.001), such as tree-in-bud sign, ground-glass opacity, consolidation, mass, and cavitation. MRI was advantageous in identifying caseation and liquefactive necrosis based on inhomogeneous signal distribution within consolidations and also in identifying mild pleural effusion. The optimized lung MRI was comparable to CT in detecting non-calcified nodules (κ=0.90), with overall sensitivity of 50.0%, 91.1%, and 100% for nodules of size <5 mm, 5-10 mm, and >10 mm, respectively. However, MRI was less effective in identifying lesions with calcification. CONCLUSIONS: The clinical implementation of an optimized MRI protocol with the MultiVane motion correction technique for imaging pulmonary TB is feasible. Lung MRI without ionizing radiation is a promising alternative to the clinical standard CT, especially for pregnant women, children, adolescents, and patients requiring short-term and repeated follow-up observations.