Towards personalized nutritional treatment for malnutrition using machine learning-based screening tools

Raphaeli, Orit; Singer, Pierre

doi:10.1016/j.clnu.2021.08.013

Cited by 13 publications

(9 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…An automatic system based on Artificial Intelligence (AI) able to compare the established quantity of food before and after consumption through images is promising in detecting patients at high risk of malnutrition during hospitalization [106]. Modeling strategies as machine learning-based algorithms can be useful in the analysis of electronic health records to improve the diagnosis and management of patients at risk of malnutrition or malnourishment [107,108].…”

Section: Nutritional Recommendationsmentioning

confidence: 99%

Malnutrition in Hospitalized Old Patients: Screening and Diagnosis, Clinical Outcomes, and Management

et al. 2022

View full text Add to dashboard Cite

Malnutrition in hospitalized patients heavily affects several clinical outcomes. The prevalence of malnutrition increases with age, comorbidities, and intensity of care in up to 90% of old populations. However, malnutrition frequently remains underdiagnosed and undertreated in the hospital. Thus, an accurate screening to identify patients at risk of malnutrition or malnourishment is determinant to elaborate a personal nutritional intervention. Several definitions of malnutrition were proposed in the last years, affecting the real frequency of nutritional disorders and the timing of intervention. Diagnosis of malnutrition needs a complete nutritional assessment, which is often challenging to perform during a hospital stay. For this purpose, various screening tools were proposed, allowing patients to be stratified according to the risk of malnutrition. The present review aims to summarize the actual evidence in terms of diagnosis, association with clinical outcomes, and management of malnutrition in a hospital setting.

show abstract

Section: Nutritional Recommendationsmentioning

confidence: 99%

Malnutrition in Hospitalized Old Patients: Screening and Diagnosis, Clinical Outcomes, and Management

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Although detection systems based on artificial intelligence have been developed in recent years [ 24 ], experts continue to recommend as a goal of personalized medicine that screening be performed by nursing staff or remotely by a health professional. It should be noted that the dynamics of including screening by default in patient care should not be exclusive to the nursing staff and can be extended to any other health care professional.…”

Section: Discussionmentioning

confidence: 99%

The Determination of a Consensus Nutritional Approach for Cancer Patients in Spain Using the Delphi Methodology

Llanos

Vera²,

Contreras-Martínez

2022

Nutrients

View full text Add to dashboard Cite

Malnutrition has a multifactorial origin and can be caused by cancer. This study determined the consensus of a panel of experts on the nutritional approach for cancer patients in Spain using a multidisciplinary approach. Using the Delphi methodology, a 74-question questionnaire was prepared and sent to 46 experts. The areas of knowledge addressed were the nutritional status of the cancer patient, nutritional screening, nutritional therapy, patient referral, and multidisciplinary care. A total of 91.7% of the experts agreed with the questions posed on nutritional status, 60.0% with those on nutritional screening, 76.7% with those on nutritional therapy, and the entire panel of experts agreed with the questions posed on patient referral and multidisciplinary care. The experts agreed upon a high prevalence of malnutrition among cancer patients in Spain. Unlike medical and radiation oncologists, medical nutrition specialists believe that body composition assessment should not be carried out in all types of cancer patients during nutritional screening and that interventions can be conducted outside the oncology clinic. In general, it is recommended that nursing staff routinely perform nutritional screening before starting cancer treatment. It is necessary to develop a multidisciplinary action protocol that includes nutritional and/or sarcopenia screening.

show abstract

“…Reasons in the literature for suboptimal processes include lack of awareness and knowledge around hospital malnutrition and screening, 32,34,35 unclear and unstructured screening and assessment processes, 32,36 lack of necessary information to make a diagnosis including weight, 37 inadequate staffing levels or human resources, 34,38,39 communication breakdowns, 40 nonstandardised or inconsistent data collection and documentation methods, difficult data analysis of dietitian workflows, 34,39,41,42 and local technological capability limitations and use of an electronic health record (EHR). 35,41,43,44 Strategies to address the barriers and inefficiencies in malnutrition identification are critical for achieving best practice in the clinical management of these at-risk patients, as well as to ensure adequate reimbursement for organisations to treat these at-risk patients through optimal clinical coding. Systematic auditing and process analysis with the use of electronic systems and tools can assist with achieving this.…”

Section: Introductionmentioning

confidence: 99%

“…Systematic auditing and process analysis with the use of electronic systems and tools can assist with achieving this. [39][40][41][43][44][45] Gaps in the established process of screening, assessment, diagnosis and coding of malnourished patients at Mater Health were identified in 2015. This reflects the findings in the literature, confirming that simply having a clear and documented malnutrition identification process in place is not sufficient for delivering optimal identification, diagnosis and coding.…”

Section: Introductionmentioning

confidence: 99%

Implementation of an electronic solution to improve malnutrition identification and support clinical best practice

McCray

Barsha

Maunder

2022

J Human Nutrition Diet

View full text Add to dashboard Cite

Background Routine malnutrition risk screening of patients is critical for optimal care and comprises part of the National Australian Hospital Standards. Identification of malnutrition also ensures reimbursement for hospitals to adequately treat these high‐risk patients. However, timely, accurate screening, assessment and coding of malnutrition remains suboptimal. The present study aimed to investigate manual and digital interventions to overcome barriers to malnutrition identification for improvements in the hospital setting. Methods Retrospective reporting on malnutrition identification processes was conducted through two stages: (1) manual auditing intervention and (2) development of a digital solution – the electronic malnutrition management solution (eMS). Repeated process audits were completed at approximately 6‐monthly intervals through both stages between 2016 and 2019 and the results were analysed. In Stage 2, time investment and staff adoption of the digital solution were measured. Results Overall, the combined effect of both regular auditing and use of the eMS resulted in statistically significant improvements across all six key measures: patients identified (97%–100%; p < 0.001), screened (68%–95%; p < 0.001), screened within 24 h (51%–89%; p < 0.001), assessed (72%–95%; p < 0.001), assessed within 24 h (66%–93%; p < 0.001) and coded (81%–100%; p = 0.017). The eMS demonstrated a reduction in screening time by over 60% with user adoption 100%. Data analytics enabled automated, real‐time auditing with a 95% reduction in time taken to audit. Conclusions A single digital solution for management of malnutrition and automation of auditing demonstrated significant improvements where manual or combinations of manual and electronic systems continue to fall short.

show abstract

Towards personalized nutritional treatment for malnutrition using machine learning-based screening tools

Cited by 13 publications

References 12 publications

Malnutrition in Hospitalized Old Patients: Screening and Diagnosis, Clinical Outcomes, and Management

Malnutrition in Hospitalized Old Patients: Screening and Diagnosis, Clinical Outcomes, and Management

The Determination of a Consensus Nutritional Approach for Cancer Patients in Spain Using the Delphi Methodology

Implementation of an electronic solution to improve malnutrition identification and support clinical best practice

Contact Info

Product

Resources

About