東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁 到查詢結果 [ null ]

切換: 標籤 | MARC模式 | ISBD

Data-Driven Approaches to Nutritiona...

Pleuss, James D.,

FindBook

Google Book

Amazon

博客來

Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment // James D Pleuss.
作者:	Pleuss, James D.,
面頁冊數:	1 electronic resource (194 pages)
附註:	Source: Dissertations Abstracts International, Volume: 86-11, Section: B.
Contained By:	Dissertations Abstracts International86-11B.
標題:	Computer science. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31846649
ISBN:	9798315736622

Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment /
Pleuss, James D.,

Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment /James D Pleuss. - 1 electronic resource (194 pages)

Source: Dissertations Abstracts International, Volume: 86-11, Section: B.

The links between nutrition and health are well documented, connecting foods and nutrients to long and short-term health outcomes. Lacking, however, are data-driven approaches that take advantage of advancements in machine learning, a particularly useful endeavor given the complex and multi-faceted nature of nutrition. In this thesis we look at how data-driven approaches to nutritional epidemiology and dietary assessment can offer a new perspective from which to view our diets, particularly during pregnancy, an under-researched stage of life.First, we focus on ultra-processed foods (UPFs), which are ubiquitous and growing in the global diet, and associated with many negative health factors. We leverage the natural hierarchy of foods and a hierarchical ranking methodology to test whether specific UPFs or categories of UPFs might improve predictions of cardiometabolic health risks. In doing so, we learn which UPFs are most important for predicting cardiometabolic risk and advocate for further research of these UPFs which might enable causal discoveries. Next, we introduce a new dataset from the Temporal Research in Eating, Nutrition, and Diet during Pregnancy (TREND-P) study which collected up to two rounds of 14-day dietary records with timestamped eating occasions from 150 pregnant individuals during their second trimester with associated biometric and health data. The study reveals significant within- and between-person variation across the micro- and macro-nutrients most critical during pregnancy. Our next paper employed the detailed timing of this dataset to develop a new chrononutrition scoring system, pairing morning and night intake scores with machine learning methods to identify associations between the time at which foods are consumed and maternal impaired glucose tolerance.We then step back to consider the validity of current methodological approaches to dietary assessment. Existing calculations to determine the required number of days of dietary assessment for nutrition studies focus on within-person variation across cohorts but fail to consider how frequently individuals meet the requirements. We capitalize on the length of the TREND-P study and the nonparametric nature of bootstrapping to provide researchers with an understanding of how their research design choices impact the accuracy of recorded intakes.Finally, nutrition research strives to establish causal relationships associated with dietary intakes, but these relationships are often limited in their generalizability to other populations. Here we introduce a new algorithm for causal discovery that goes beyond causal relationship identification. It incorporates static characteristics (age, education, comorbidities, etc.) to determine the context within which a causal relationship applies or whether the context strengthens or weakens that causal relationship. We validate our algorithm against benchmarks and apply it to physical activity and nutrition data. This thesis covers a wide range of data science initiatives (feature selection, bootstrapping, causal inference), applying them to advancing areas of nutrition research (UPFs, chrononutrition, diet/physical activity relationships), while providing a new dataset (TREND-P) that is suited for complex machine learning tasks. Our hopes is that this research will spark further data science advances in nutritional epidemiology.

English

ISBN: 9798315736622Subjects--Topical Terms:

523869
Computer science.
Subjects--Index Terms:

Data-driven approaches

Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment /
LDR:04912nmm a22004573i 4500 001 2437170
005 20251118105039.5
006 m o d
007 cr|nu||||||||
008 261011s2025 miu||||||m |||||||eng d
020 $a 9798315736622
035 $a (MiAaPQD)AAI31846649
035 $a AAI31846649
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Pleuss, James D., $e author. $3 3828068
245 1 0 $a Data-Driven Approaches to Nutritional Epidemiology and Dietary Assessment / $c James D Pleuss.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2025
300 $a 1 electronic resource (194 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 86-11, Section: B.
500 $a Advisors: Kleinberg, Samantha Committee members: Wang, Ping; Asan, Onur; Oh, Hyewon; Deierlein, Andrea.
502 $b Ph.D. $c Stevens Institute of Technology $d 2025.
520 $a The links between nutrition and health are well documented, connecting foods and nutrients to long and short-term health outcomes. Lacking, however, are data-driven approaches that take advantage of advancements in machine learning, a particularly useful endeavor given the complex and multi-faceted nature of nutrition. In this thesis we look at how data-driven approaches to nutritional epidemiology and dietary assessment can offer a new perspective from which to view our diets, particularly during pregnancy, an under-researched stage of life.First, we focus on ultra-processed foods (UPFs), which are ubiquitous and growing in the global diet, and associated with many negative health factors. We leverage the natural hierarchy of foods and a hierarchical ranking methodology to test whether specific UPFs or categories of UPFs might improve predictions of cardiometabolic health risks. In doing so, we learn which UPFs are most important for predicting cardiometabolic risk and advocate for further research of these UPFs which might enable causal discoveries. Next, we introduce a new dataset from the Temporal Research in Eating, Nutrition, and Diet during Pregnancy (TREND-P) study which collected up to two rounds of 14-day dietary records with timestamped eating occasions from 150 pregnant individuals during their second trimester with associated biometric and health data. The study reveals significant within- and between-person variation across the micro- and macro-nutrients most critical during pregnancy. Our next paper employed the detailed timing of this dataset to develop a new chrononutrition scoring system, pairing morning and night intake scores with machine learning methods to identify associations between the time at which foods are consumed and maternal impaired glucose tolerance.We then step back to consider the validity of current methodological approaches to dietary assessment. Existing calculations to determine the required number of days of dietary assessment for nutrition studies focus on within-person variation across cohorts but fail to consider how frequently individuals meet the requirements. We capitalize on the length of the TREND-P study and the nonparametric nature of bootstrapping to provide researchers with an understanding of how their research design choices impact the accuracy of recorded intakes.Finally, nutrition research strives to establish causal relationships associated with dietary intakes, but these relationships are often limited in their generalizability to other populations. Here we introduce a new algorithm for causal discovery that goes beyond causal relationship identification. It incorporates static characteristics (age, education, comorbidities, etc.) to determine the context within which a causal relationship applies or whether the context strengthens or weakens that causal relationship. We validate our algorithm against benchmarks and apply it to physical activity and nutrition data. This thesis covers a wide range of data science initiatives (feature selection, bootstrapping, causal inference), applying them to advancing areas of nutrition research (UPFs, chrononutrition, diet/physical activity relationships), while providing a new dataset (TREND-P) that is suited for complex machine learning tasks. Our hopes is that this research will spark further data science advances in nutritional epidemiology.
546 $a English
590 $a School code: 0733
650 4 $a Computer science. $3 523869
650 4 $a Bioinformatics. $3 553671
650 4 $a Epidemiology. $3 568544
650 4 $a Obstetrics. $3 634501
650 4 $a Nutrition. $3 517777
653 $a Data-driven approaches
653 $a Ultra-processed foods
653 $a Pregnancy
653 $a Glucose tolerance
653 $a Chrononutrition
690 $a 0984
690 $a 0766
690 $a 0380
690 $a 0715
690 $a 0570
710 2 $a Stevens Institute of Technology. $b Computer Science. $e degree granting institution. $3 3818553
720 1 $a Kleinberg, Samantha $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 86-11B.
790 $a 0733
791 $a Ph.D.
792 $a 2025
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31846649