Progress 03/01/23 to 02/29/24
Outputs
Target Audience:The current research presents a machine learning (ML) framework in a sub-area of food science. The project has introduced students to the application of ML tools. Some research results have already been introduced in the curriculum and shared with the broader community. The project will enable students to pursue more classes and training in ML applications across various agriculture projects and research. In the second year of our project, we expanded our target audience to include new collaborators from academia and student trainees (2 graduate students and two undergraduates). The group published six manuscripts in peer-reviewed journals and attended three conferences. Student participants attended conferences that provide them access to a diverse range of sessions, workshops, and presentations delivered by industry experts. By attending these sessions, students were able to enhance their knowledge and stay updated on the latest trends, research, and best practices in their field. The conferences offered a platform to connect and network with peers, colleagues, and professionals from various organizations. Building new relationships and expanding professional networks can lead to collaborations, mentorship opportunities, and future career prospects. The conferences included hands-on workshops and training sessions that focus on developing specific skills or learning new tools and techniques in AI and Data Science. These interactive sessions helped the project team and students to acquire practical skills that can be applied in their work. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? Publication Support: Guidance and assistance were provided in editing and structuring research publications, ensuring that they met academic writing standards. The objective was to enhance clarity, coherence, and readability, thereby helping students effectively communicate their research findings. Computational Technology Utilization: Assistance was provided for utilizing computational tools relevant to data science projects. This facilitated students' proficiency in data analysis and visualization, enabling them to effectively analyze and interpret complex data sets. Proofreading and Editing: Proofreading services were offered to students, providing them with constructive feedback on grammar, syntax, and structure. This helped in enhancing the quality and coherence of their written work. Additionally, we reviewed and refined technical content to ensure accuracy and precision in conveying data science concepts. Impact: Empowered students with the essential skills and knowledge crucial for effective publication writing, computational analysis, and machine learning implementation. Guidance and support enhanced students' confidence in utilizing computational tools and algorithms, fostering a deeper understanding of data science methodologies. Additionally, we facilitated the improved communication of data science concepts by providing effective proofreading and terminology clarification. Conclusion: Through our assistance and guidance, we were able to enhance the students' capabilities in various aspects of data science research and publication. Our objective was to equip them with essential skills in publication writing, computational technology, machine learning algorithms, proofreading, and terminology, thereby strengthening their proficiency and readiness to succeed in their data science endeavors. The conferences bring together professionals from different backgrounds and perspectives. Attending presentations and discussions can expose individuals to innovative ideas, approaches, and solutions that they can apply to their own work. Participants had the opportunity to present their own work through poster sessions, oral presentations, or panel discussions. These hands-on activities improved participants presentation skills, boost confidence, and provide valuable feedback from peers and experts. The participants engaged in presenting research findings and leading discussions that enhanced their skills and professional visibility within their work. It helped establish credibility and recognition as a thought leader in the field. The conferences served as a platform for finding potential collaborators and partners for research projects, initiatives, or business ventures. Building connections with like-minded individuals can lead to fruitful collaborations in the future. How have the results been disseminated to communities of interest?We have disseminated the results of our work through: (1) six peer-reviewed papers published in reputable journals and (2) three public presentations to a diverse audience at professional conferences over this project period, describing the aims of integrating data science techniques in food science research and what we have achieved so far. The project team have provided opportunities for connecting our research to some middle and high school students at the Early College High in Dover. One of the CO-PIs presented to environment engineering students in a seminar, sharing the project results. The seminar shows that the ideas presented in the project can be replicated in other subject areas, including environmental engineering and science. What do you plan to do during the next reporting period to accomplish the goals?We plan to convene two speaking opportunities on novel uses of ML/AI in food system. We also plan to on-board another graduate student and an undergraduate who will be partially supported with this grant. We will also include the reviews of the evaluator to refine initial plans for future grant proposals and summarize the ML/AI areas explored which are critically needed to advance Food science research and application of AI for sustainable product development. We will continue with the development and refining of curriculum modules for high school and undergraduates in various modalities. We are also planning to validate the results using out-of-sample data.
Impacts
What was accomplished under these goals?
The project collaborated across three main themes including 1) determining the ability of ML algorithms including artificial neural networks (ANNs), decision trees (DTs), k-nearest neighbors (KNNs), and support vector machines (SVMs) to assess the variability in fatty classes (SFA, MUFA, and PUFA) in US- snacks consumed over a selected period. These approaches proved to be time-efficient and cost-effective to predict the nutritional value of the snacks. 2). assessing the combined utility of ATR-FTIR spectroscopy and ML techniques to identify and classify pure njangsa seed oil (NSO), palm kernel oil (PKO), coconut oil (CCO), njangsa seed-palm kernel oil (NSOPKO) and njangsa seed-coconut oil (NSOCCO) margarine. Additionally, it strove to quantify the degree of adulteration in each oil and margarine using ML regression models and sunflower oil and canola-flax seed oil margarine as adulterants. This combined use of FTIR spectroscopy and ML techniques to create models and demonstrated the qualitative classification of pure oils and predictions of adulteration in oils and margarine. PCA was integrated with the ML methods to increase classification accuracy for pure and adulterated oils and margarines by selecting the features that described most of their variance and summarized them into two PCs, which ensured efficient sample segregation. The FTIR spectroscopy technique avoided the need for laborious and complicated sample preparation making it a rapid and simple method for analysis. The demonstrated fingerprinting method suggests that ML methods in conjunction with FTIR spectroscopy can reliably classify and quantify adulterants in oil and margarine and could be further improved for applications in quality control settings to quickly authenticate new products, and 3). Predicting the quality of foods and beverages formulated with plant-based ingredients using Nutri-score and ML techniques. ML techniques were used to connect different datasets to insights about the quality of plant-based foods. Faster determination of these nutrients in foods through these models could promote intervention strategies by regulatory bodies to generate new or combined ingredients which can minimize calorie intake from snacks by consumers. It will increase awareness of the healthiness of different foods and cater to consumers' demand for personalized nutrition. Deep learning concepts could be developed for other foods that rely on tedious analytical/instrumentation methods to save time and minimize waste. Collaborating with a multidisciplinary team on implementing ML and AI techniques and the availability of a wide range of open-source dataset on food quality and processing to be evaluated ML techniques for developing predictive models for food product developers and producers. This process was guided by our multi-disciplinary leadership team who provided guidance on strategy, knowledge elicitation, and data collection needs, data gathering, gap analysis, and pivot decision point. One graduate student thesis supervised on integrating ML techniques on primary and secondary data on food quality and food authentication. The students used mined data from the National Health and Nutrition Survey (NHANES) and data collected from a study using FTIR spectroscopy to develop ML models. The project results provide results that have a significant social impact. The approach can replicate the look of different foods and address health-related issues connected to the product's contents. For instance, this approach is more critical for low-income areas where the population has little knowledge of the food they are consuming. Health professionals can also use the current approach in addressing nutrition guidance for different populations.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Tachie, C., Nwachukwu, I.D. and Aryee, A.N.A. (2023). Trends, and innovations in the formulation of plant-based foods. Food Production Processing and Nutrition, 5, 16; https://doi.org/10.1186/s43014-023-00129-0
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Tachie, C., Tawiah, N.A. and Aryee, A.N.A. (2023). Using machine learning models to predict the quality of plant-based foods. Current Research in Food Science, 7: 100544; https://doi.org/10.1016/j.crfs.2023.100544
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Tachie, C.Y.E., Obiri-Ananey, D., Tawiah, N.A., Attoh-Okine, N. and Aryee, A.N.A. (2023). Machine learning approaches for predicting fatty acid classes in popular US snacks using NHANES data. Nutrients, 15(15): 3310; https://doi.org/10.3390/nu15153310
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Wei, Z., Ananga, A., Ukuku D.O. and Aryee A.N.A. (2023). High salt concentration affects the microbial diversity of cassava during fermentation as revealed by 16S rRNA gene sequencing. Fermentation, 9(8): 727; https://doi.org/10.3390/fermentation9080727
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Tachie, C.Y.E., Obiri-Ananey, D., Alfaro-Cordoba, M., Tawiah, N.A. and Aryee, A.N.A. (2024). Classification of oils and margarines by FTIR spectroscopy in tandem with machine learning. Food Chemistry, 431: 137077; https://doi.org/10.1016/j.foodchem.2023.137077
- Type:
Journal Articles
Status:
Accepted
Year Published:
2024
Citation:
Tachie, C., Onuh, J.O. and Aryee, A.N.A. Nutritional and potential health benefits of fermented proteins. Journal of the Science of Food and Agriculture, 104(3): 1223-1233; https://doi.org/10.1002/jsfa.13001
|
Progress 03/01/22 to 02/28/23
Outputs
Target Audience:The project targets underrepresented and underserved communities and majors/disciplinesnot traditionally associated with data science. The project provided science-based knowledge duringclassroom instruction, laboratory instruction, or practicum experiences; development of curriculum or innovative teaching methodologies; internships; workshops; experiential learning opportunities; and outreach. Changes/Problems:A major problems was the delay in on-boarding studentsdue to COVID-19. What opportunities for training and professional development has the project provided?Training activities: Both undergraduates and graduate students are benefitting from with faculty and collaborators serving as mentors at various stages of the project. Professional developmentactivities: Faculty has participarted inworkshops, seminars, study groups, and individual study to increase knowledge such as theSMART-DART: Health Equity:SupportingMinority andRegionalTraining inData &AI forResearchers ofTomorrow: Health Equity Cohort. How have the results been disseminated to communities of interest?The results has been disseminated through presentations and manuscripts. What do you plan to do during the next reporting period to accomplish the goals?We plan torevist the regression analysisusing the results from both the EDA and some of the ML algorithms,implementnovel classification scheme and categorical variables encoding of open-sourced data on food and nutrition. We also plan to complete and submit other manuscripts for peer-review.
Impacts
What was accomplished under these goals?
We have incoporated machine learning techniques in studenttraining and provided faculty opportunities to develop data science approaches for food science research through training. Exploratory Data Analysis (EDA) with experimental and open-source data (National Health and Nutrition Survey (NHANES)) data Regression analysis of selected variables performed. Prior to performing the analysis, the covariates were carefully selected based on their significance. Multi-class classification system for commonly consumed snacks in the US using ML algorithms such as support vector machine (SVM), decision tree (DT), light gradient-boosting machine (LightGBM), K-nearest neighbor (KNN), logistic regression (LR), random forest (RF), artificial neural networks (ANN) to predict the amount of the fatty acid classes in these snacks K- cross-validation to train and test data sets Model complex non-linear data sets by incorporating interactions between sparse matrices and nutritional variables like fatty acids and snacks to find non-linear relationships between the outcomes that conventional regression models might miss Assessing discrimination between authentic and adulterated oils and margarines using a combination of FTIR and chemometric (machine learning algorithms)
Publications
- Type:
Conference Papers and Presentations
Status:
Submitted
Year Published:
2023
Citation:
Tachie, C., Attoh-Okine, N.O., Alfaro-C�rdoba, M. and Aryee, A.N.A. Application of FTIR spectroscopy and chemometrics for the authentication of oils and margarines (Submitted for presentation at the 2023 AOCS Annual Meeting & Expo 01/13/2023)
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2023
Citation:
Aryee, A.N.A. Multidisciplinary opportunities in Food Science Research. Tuskegee University, Seminar in Food and Nutritional Sciences, January 25, 2023
- Type:
Journal Articles
Status:
Awaiting Publication
Year Published:
2023
Citation:
Tachie, C., Nwachukwu, I.D. and Aryee, A.N.A. Trends and innovations in the formulation of plant-based foods. Food Production, Processing and Nutrition. DOI: 10.1186/s43014-023-00129-0 (Accepted 12/26/2022)
- Type:
Journal Articles
Status:
Submitted
Year Published:
2023
Citation:
Tachie, C., Attoh-Okine, N.O., Tawiah, N.A. and Aryee, A.N.A. Predicting Fatty Acid Classes in Popular US Snacks Using NHANES Data and Machine Learning Approaches (Submitted to Bioengineered 12/07/2022)
- Type:
Journal Articles
Status:
Under Review
Year Published:
2023
Citation:
Tachie, C., Attoh-Okine, N.O., Tawiah, N.A. and Aryee, A.N.A. Combination of FTIR spectroscopy and machine learning for non-destructive product identification (To be submitted to Journal of Food Control)
|
Progress 03/01/21 to 02/28/22
Outputs
Target Audience:For the first period of performance of this grant, which is being implemented here at DSU and UD, we have submitted an abstract to the Association of Research Directors (ARD) and a review manuscript is in preparation. Changes/Problems:The project has enhanced/intensified its linkage with our partner at UD and other faculty are now expressing interest. What opportunities for training and professional development has the project provided? The idea of using Data Science and Machine Learning to Food Science was presented to undergraduate students at the College of Agriculture at the University of Delaware A graduate student is developing a GitHub Site for Data Science and ML application in Food Science How have the results been disseminated to communities of interest? Preliminary results were presented at the School of Agriculture, UD Some results with be presented during ARD What do you plan to do during the next reporting period to accomplish the goals? Additional data will be collected from food databases and the literature Additional Exploratory Data Analysis to e preformed on data On-board another graduate student
Impacts
What was accomplished under these goals?
Data was collected from food databases and the literature Exploratory Data Analysis was completed on the data Preliminarily remarks were presented based on the results Graduate student at UD presented the preliminary results during Graduate Seminar
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Asuzu, P.C., Trompeter, N.S., Cooper, C.R., Besong, S.A. and Aryee, A.N.A. (2022). Cell culture-based assessment of toxicity and therapeutics of phytochemical antioxidants. Molecules 2022, 27, 1087.
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Aryee, A.N.A., Akanbi, T.O., Nwachukwu, I.D. and Gunathilake, T. (2022). Perspectives on preserving lipid quality and strategies for value enhancement. Current Opinion in Food Science, 44: 100802.
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
Pre-processing Treatments Improved the Physicochemical Properties of Bambara Groundnut Flours and Preference of Formulated Cake
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2022
Citation:
Enhancing Oxidative Stability and Delivery of Njangsa (Ricinodendron heudelotii) Seed Oil by Encapsulation
- Type:
Conference Papers and Presentations
Status:
Accepted
Year Published:
2022
Citation:
Machine Learning Approaches to Predict Micronutrients Content in Plant-based Foods
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
Profiling Carotenoids and Other Bioactives in Selected Starchy Staples
- Type:
Journal Articles
Status:
Under Review
Year Published:
2022
Citation:
In vitro Assessment of Efficacy and Cytotoxicity of Prunus africana Extracts on Prostate Cancer C4-2 Cells
|