Development and evaluation of a deep learning framework for the diagnosis of malnutrition using a 3D facial points cloud: A cross-sectional study.
The feasibility of diagnosing malnutrition using facial features has been validated. A tool to integrate all facial features associated with malnutrition for disease screening is still demanded. This work aims to develop and evaluate a deep learning (DL) framework to accurately determine malnutrition based on a 3D facial points cloud.
A group of 482 patients were studied in this perspective work. The 3D facial data were obtained using a 3D camera and represented as a 3D facial points cloud. A DL model, PointNet++, was trained and evaluated using the points cloud as inputs and classified the malnutrition states. The performance was evaluated with the area under the receiver operating characteristic curve, accuracy, specificity, sensitivity, and F1 score.
Among the 482 patients, 150 patients (31.1%) were diagnosed as having moderate malnutrition and 54 patients (11.2%) as having severe malnutrition. The DL model achieved the performance with an area under the receiver operating characteristic curve of 0.7240 ± 0.0416.
The DL model achieved encouraging performance in accurately classifying nutrition states based on a points cloud of 3D facial information of patients with malnutrition.
Wang X
，Liu Y
，Rong Z
，Wang W
，Han M
，Chen M
，Fu J
，Chong Y
，Long X
，Tang Y
，Chen W
... -
《-》
Determining the optimal value of the Geriatric Nutritional Risk Index to screen older patients with malnutrition risk: A study at a university hospital in Japan.
The Geriatric Nutritional Risk Index (GNRI) can predict nutritional risk. However, just a few studies have validated the optimal cut-off value of GNRI for nutrition screening in older patients. Hence, this study aimed to determine the optimal value of GNRI to screen the risk of malnutrition among older patients.
This retrospective cross-sectional study was carried out with 5867 consecutive older adult patients who were admitted to an academic hospital in Japan. Receiver operating characteristic curve analyses were carried out to obtain the optimal cut-off value of GNRI, and the results were compared against the Mini Nutritional Assessment - Short Form and Malnutrition Universal Screening Tool. The validation of the obtained cut-off value was examined on the concordance rate of malnutrition diagnosis based on the European Society of Clinical Nutrition and Metabolism criteria.
The mean age of the patients was 76.0 ± 7.0 years. The optimal cut-off value of GNRI for Mini Nutritional Assessment - Short Form ≤11 points was 95.92 (area under the curve 0.827 [0.817-0.838], P < 0.001), and that for Malnutrition Universal Screening Tool ≥1 point was 95.95 (area under the curve 0.788 [0.776-0.799], P < 0.001). By adapting GNRI <96 points as an initial screening cut-off in the European Society of Clinical Nutrition and Metabolism-defined malnutrition process, the concordance rates of comparisons were 98.5% and 98.5% for Mini Nutritional Assessment - Short Form-based and MUST-based diagnosis, respectively.
The study showed GNRI <96 points as the optimal cut-off value for nutritional screening. GNRI might be one of the easy-to-use tools for nutritional screening and for diagnosing malnutrition in older adults. Geriatr Gerontol Int 2020; 20: 811-816.
Ishida Y
，Maeda K
，Nonogaki T
，Shimizu A
，Yamanaka Y
，Matsuyama R
，Kato R
，Ueshima J
，Mori N
... -
《-》
Automatic three-dimensional facial symmetry reference plane construction based on facial planar reflective symmetry net.
Three-dimensional (3D) facial symmetry analysis is based on the 3D symmetry reference plane (SRP). Artificial intelligence (AI) is widely used in the dental and oral sciences. This study developed a novel deep learning model called the facial planar reflective symmetry net (FPRS-Net) to automatically construct an SRP and established a method for defining a 3D point-cloud region of interest (ROI) and high-dimensional feature computations suitable for this network model.
Overall, 240 patients were enroled. The deep learning model was trained and predicted using 200 samples, and its clinical suitability was evaluated with 40 samples. Four FPRS-Net models were prepared, each using supervised and unsupervised learning approaches based on full facial and ROI data (FPRS-NetS, FPRS-NetSR, FPRS-NetU, and FPRS-NetUR). These models were trained on 160 3D facial datasets, validated on 20 cases, and tested on another 20 cases. The model predictions were evaluated using an additional 40 clinical 3D facial datasets by comparing the mean square error of the SRP between the parameters predicted by the four FPRS-Net models and the truth plane. The clinical suitability of FPRS-Net models was evaluated by measuring the angle error between the predicted and ground-truth planes; experts evaluated the predicted SRP of the four FPRS-Net models using the visual analogue scales (VAS) method.
The FPRS-NetSR and FPRS-NetU models achieved an average angle error of 0.84° and 0.99° in predicting 3D facial SRP, respectively, with a VAS value of >8. Using the four FPRS-Net models to create an SRP in 40 cases of 3D facial data required <4 s.
Our study demonstrated a new solution for automatically constructing oral clinical 3D facial SRPs.
This study proposes a novel deep learning algorithm (FPRS-Net) to construct a symmetry reference plane that can reduce workload, shorten the time required for digital design, reduce dependence on expert experience, and improve therapeutic efficiency and effectiveness in dental clinics.
Zhu Y
，Zhang L
，Liu S
，Wen A
，Gao Z
，Qin Q
，Gao L
，Zhao Y
，Wang Y
... -
《-》
Geriatric Nutrition Risk Index is comparable to the mini nutritional assessment for assessing nutritional status in elderly hospitalized patients.
Malnutrition is common among hospitalized elderly patients, and the prevalence is increasing not only in Malaysia but also in the rest of the world. The Geriatric Nutrition Risk Index (GNRI) and the Mini Nutritional Assessment (MNA) were developed to identify malnourished individuals among this group. The MNA was validated as a nutritional assessment tool for the elderly. The GNRI is simpler and more efficient than the MNA, but studies on the use of the GNRI and its validity among the Malaysian population are absent. This study aimed to determine the prevalence of malnourished hospitalized elderly patients and assess the criterion validity of the GNRI and MNA among the geriatric Malaysian population against the reference standard for malnutrition, the Subjective Global Assessment (SGA), and determine whether the optimal cutoff value of the GNRI is suitable for the Malaysian population and determine the optimal tool for use in this population.
A cross-sectional study was conducted among 134 geriatric patients with a mean age of 68.9 ± 8.4 who stayed at acute care wards in Hospital Tuanku Ampuan Rahimah, Klang from July 2017 to August 2017. The SGA, MNA, and GNRI were administered through face-to-face interviews with all the participants who gave their consent. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the GNRI and MNA were analyzed against the SGA. Receiver-operating characteristic (ROC) curve analysis was used to obtain the area under the curve (AUC) and suitable optimal cutoff values for both the GNRI and MNA.
According to the SGA, MNA, and GNRI, 26.9%, 42.5%, and 44.0% of the participants were malnourished, respectively. The sensitivity, specificity, PPV, and NPV for the GNRI were 0.622, 0.977, 0.982, and 0.558, respectively, while those for the MNA were 0.611, 0.909, 0.932, and 0.533, respectively. The AUC of the GNRI was comparable to that of the MNA (0.831 and 0.898, respectively). Moreover, the optimal malnutrition cutoff value for the GNRI was 94.95.
The prevalence of malnutrition remains high among hospitalized elderly patients. Validity of the GNRI is comparable to that of the MNA, and use of the GNRI to assess the nutritional status of this group is proposed with the new suggested cutoff value (GNRI ≤ 94.95), as it is simpler and more efficient. Underdiagnosis of malnutrition can be prevented, possibly reducing the prevalence of malnourished hospitalized elderly patients and improving the quality of the nutritional care process practiced in Malaysia.
Abd Aziz NAS
，Mohd Fahmi Teng NI
，Kamarul Zaman M
《Clinical Nutrition ESPEN》
ResearchGate
(全网免费下载)
钛学术
(全网免费下载)
