Combining stable isotopes and multi-elements with machine learning chemometric models to identify the geographical origins of Tetrastigma hemsleyanum Diels et Gilg DOI
Lu Bai, Zixuan Zhang, Yalan Li

et al.

Food Chemistry, Journal Year: 2024, Volume and Issue: 469, P. 142496 - 142496

Published: Dec. 24, 2024

Language: Английский

Digital intelligence technology: new quality productivity for precision traditional Chinese medicine DOI Creative Commons
Junqing Zhu, Xiaonan L. Liu, Peng Gao

et al.

Frontiers in Pharmacology, Journal Year: 2025, Volume and Issue: 16

Published: April 8, 2025

Traditional Chinese medicine is a complex medical system characterized by multiple metabolites, targets, and pathways, known for its low drug resistance significant efficacy. However, challenges persist within medicine, including difficulties in assessing the quality of Botanical drugs, reliance on experiential knowledge disease diagnosis treatment, lack clarity regarding pharmacological mechanisms medicine. The advancement digital intelligence technology driving shift towards precision model. This transition propels into an era precision, intelligence, digitalization. paper introduces standard technologies explores application control evaluation studies research status assisting diagnosis, treatment prevention diseases, further promotes development field

Language: Английский

Citations

0
Assessing the ecotoxicological risk of nicosulfuron on maize using multi-source phenotype data and hyperspectral imaging DOI Creative Commons

Tianpu Xiao,

Yang Li,

Xiantao He

et al.

Ecotoxicology and Environmental Safety, Journal Year: 2025, Volume and Issue: 295, P. 118176 - 118176

Published: April 1, 2025

Herbicide-induced toxicity in maize crops poses significant challenges for agricultural management. Traditional assessment methods herbicide often show inconsistent accuracy. This study explores rapid and non-invasive techniques evaluating toxicity, focusing on the physiological, biochemical, growth responses of varieties subjected to two concentrations nicosulfuron. We developed a comprehensive evaluation model classify samples into three levels, showing strong correlation (r = 0.95) with traditional tassel stage assessments. Additionally, we used hyperspectral imaging coupled deep learning predict early levels following exposure. After 4 days treatment, our ToxicNet using spectral data achieved an impressive 89.66 % accuracy predicting nicosulfuron facilitating detection. Furthermore, by integrating leaf data, Soil-Plant Analysis Development (SPAD) values water content, ToxicNet-MS remarkable prediction 91.38 %. Notably, this demonstrated robust generalization across different years planting seasons, accuracies 83.33 89.89 %, respectively. These results significantly outperformed machine (Support Vector Machine, Random Forest), classical models (Multilayer Perceptron, AlexNet), spectral-based model. advancement offers promising, early, solution assessing herbicide-induced crops, ultimately benefiting both sustainable practices effective crop

Language: Английский

Citations

0
Advancements in artificial intelligence-based technologies for PFAS detection, monitoring, and management DOI
Jungsu Park, Jong-Hyun Baik,

Samuel Adjei-Nimoh

et al.

The Science of The Total Environment, Journal Year: 2025, Volume and Issue: 980, P. 179536 - 179536

Published: April 30, 2025

Language: Английский

Citations

0
Spectrum is a picture: Feasibility study of two-dimensional convolutional neural networks in spectral processing DOI
Vladislav Deev, В. В. Панчук, Ekaterina Boichenko

et al.

Microchemical Journal, Journal Year: 2024, Volume and Issue: 205, P. 111329 - 111329

Published: Aug. 3, 2024

Language: Английский

Citations

2
Analysis of Chemical Changes during Maturation of Amomum tsao-ko Based on GC-MS, FT-NIR, and FT-MIR DOI Creative Commons
Gang He,

Shaobing Yang,

Yuanzhong Wang

et al.

ACS Omega, Journal Year: 2024, Volume and Issue: 9(27), P. 29857 - 29869

Published: June 24, 2024

Amomum tsao-ko Crevost et Lemaire (A. tsao-ko) is widely grown for its high nutritional and economic value. However, the lack of a scientific harvesting quality control system has resulted in an uneven product quality. The present study was based on A. from four maturity stages same growing area, chemical trends were evaluated using combination agronomic trait analysis, spectroscopy, chromatography, chemometrics, network pharmacology. results showed that phenotypically dominant October. Spectroscopy absorbance intensity at different trend October > September August July. Further differences between found by chromatography to originate mainly alcohol, aromatic, acids, esters, hydrocarbons, ketone, heterocyclic, aldehydes. pharmacology active ingredient treatment obesity had levels this provide new idea comprehensive evaluation theoretical basis resource utilization tsao-ko.

Language: Английский

Citations

1
The effect of the fatty acid composition of fried oil on the oil-absorbing capacity of tofu puffs DOI Creative Commons
Chang Liu, Ning Wang,

Dandan Wu

et al.

LWT, Journal Year: 2024, Volume and Issue: 203, P. 116406 - 116406

Published: June 25, 2024

The influence of the fatty acid composition fried oil on oil-absorbing capacity tofu puffs was examined. Two edible vegetable oils and three blends different oleic linoleic ratios were analyzed before after frying, along with rheological properties, composition, value, peroxide p-anisidine value. As well as moisture content, microstructure changes puffs, evaluated. results revealed that viscosity all increased short-time thermal exposure, while value also increased. blend oil, an ratio 1:1, demonstrated good oxidation stability low viscosity. this had lower content weak adhesion, attributed to their microstructure. development special frying based provides a theoretical basis for production low-oil food.

Language: Английский

Citations

1
Spectrum is a Picture: Feasibility Study of Two-Dimensional Convolutional Neural Networks in Spectral Processing DOI
Vladislav Deev,

Vitaliy Panchuk,

Ekaterina Boichenko

et al.

Published: Jan. 1, 2024

In spectral data processing, a spectrum is usually represented in form of numeric vector for further processing. The same approach has been used also treatment by convolutional neural networks (CNN), initially purposed, however, image Analyzing as two-dimensional picture rather than one-dimensional potentially can improve the accuracy regression and classification models. purpose this work was to test assumption. We explored potential 2D-CNN with two case studies: Mössbauer - predict parameters numerically using (.bmp) files spectra near-infrared biological tissues. compared performance CNN types input data: pictures vectors. results indicate that proposed be helpful certain cases.

Language: Английский

Citations

0
Determination of the Geographical Origin for Tetrastigma Hemsleyanum Diels Et Gilg Based on Stable Isotopes and Multi-Elements DOI
Lu Bai, Zixuan Zhang, Yalan Li

et al.

Published: Jan. 1, 2024

Language: Английский

Citations

0
Smartphone Camera-Based Image Scanning Densitometry: Past, Present, and Future Perspectives DOI

Vidhi Vashi,

Chandni Chandarana

Journal of Analytical Chemistry, Journal Year: 2024, Volume and Issue: 79(7), P. 883 - 900

Published: July 1, 2024

Language: Английский

Citations

0
Recommendation Systems in Nutrition Field: A Scoping Review (Preprint) DOI Creative Commons
Kai Zhao,

Xinyu Xue,

Ningsu Chen

et al.

Published: July 19, 2024

BACKGROUND Recommendation systems (RS) have been widely used in the field of nutrition to promote nutritional self-management, but few NRSs adopted due various reasons. Limited studies reviewed RSs food, with some methodological flaws including limited databases searches, high heterogeneity among included and rapidly evolving nature evidence. OBJECTIVE We conducted a scoping review summarize currently available recommendation applied (NRS) published articles, patents application software explore potential gaps between development implementation. METHODS comprehensive search seven bibliographic databases, two patent four mobile apps store three websites engines for this reviews. Data extraction was by reviewers, pilot study performed before formal extraction, interrater agreement percentage needed be >75%. Discrepancies were resolved consensus or involvement third reviewer. Frequency count narrative summaries each study. RESULTS A total 877 half them released after 2022 (n=423, 48.2%) 155 (17.7%) from China. The most users overweight obese population (n=152, 17.3%), primary inputs being self-reported data on status, diet, exercise (the same n=157, 17.9% every one), output plans (n=254, 29.0%), main audience general (n=244, 27.8%). Of 49 journals essays, researchers filed reported (n=4, 3.6%), public survey (n=22, 46.8%). Forty evaluation stage, incomplete processes lack outcome. 18 artificial intelligence technologies studies, could automatically update themselves, proposed last decade. In addition, algorithms identified, only one latest knowledge-based algorithm that can improve precise matching. CONCLUSIONS While has primarily focused population, there is growing demand professional tailored special populations incorporate dynamic updates enhanced individual identification. Standardized based their technical performance clinical impact effectively support future. CLINICALTRIAL protocol registered Open Science Framework (https://doi.org/10.17605/OSF.IO/VF7NB)

Language: Английский

Citations

0