Feed-forward neural network model for hunger and satiety related VAS score prediction

Artificial intelligence for skin permeability prediction: deep learning

BACKGROUND AND OBJECTIVE

Researchers have put in significant laboratory time and effort in measuring the permeability coefficient (Kp) of xenobiotics. To develop alternative approaches to this labour-intensive procedure, predictive models have been employed by scientists to describe the transport of xenobiotics across the skin. Most quantitative structure-permeability relationship (QSPR) models are derived statistically from experimental data. Recently, artificial intelligence-based computational drug delivery has attracted tremendous interest. Deep learning is an umbrella term for machine-learning algorithms consisting of deep neural networks (DNNs). Distinct network architectures, like convolutional neural networks (CNNs), feedforward neural networks (FNNs), and recurrent neural networks (RNNs), can be employed for prediction.

METHODS

In this project, we used a convolutional neural network, feedforward neural network, and recurrent neural network to predict skin permeability coefficients from a publicly available database reported by Cheruvu et al. The dataset contains 476 records of 145 chemicals, xenobiotics, and pharmaceuticals, administered on the human epidermis in vitro from aqueous solutions of constant concentration either saturated in infinite dose quantities or diluted. All the computations were conducted with Python under Anaconda and Jupyterlab environment after importing the required Python, Keras, and Tensorflow modules.

RESULTS

We used a convolutional neural network, feedforward neural network, and recurrent neural network to predict log kp.

CONCLUSION

This research work shows that deep learning networks can be successfully used to digitally screen and predict the skin permeability of xenobiotics.

Development and validation of prediction model for early warning of ovarian metastasis risk of endometrial carcinoma

Ovarian metastasis of endometrial carcinoma (EC) patients not only affects the decision of the surgeon, but also has a fatal impact on the fertility and prognosis of patients. This study aimed build a prediction model of ovarian metastasis of EC based on machine learning algorithm for clinical diagnosis and treatment management guidance. We retrospectively collected 536 EC patients treated in Hubei Cancer Hospital from January 2017 to October 2022 and 487 EC patients from Tongji Hospital (January 2017 to December 2020) as an external validation queue. The random forest model, gradient elevator model, support vector machine model, artificial neural network model (ANNM), and decision tree model were used to build ovarian metastasis prediction model for EC patients. The predictive efficacy of 5 machine learning models was evaluated by receiver operating characteristic curve and decision curve analysis. For screening of candidate predictors of ovarian metastasis of EC, the degree of tumor differentiation, lymph node metastasis, CA125, HE4, Alb, LH can be used as a potential predictor of ovarian metastasis prediction model in EC patients. The effectiveness of the prediction model constructed by the 5 machine learning algorithms was between (area under curve [AUC]: 0.729, 95% confidence interval [CI]: 0.674-0.784) and (AUC: 0.899, 95% CI: 0.844-0.954) in the training set and internal verification set, respectively. Among them, the ANNM was equipped with the best prediction effectiveness (training set: AUC: 0.899, 95% CI: 0.844-0.954) and (internal verification set: AUC: 0.892, 95% CI: 0.837-0.947). The prediction model of ovarian metastasis of EC patients based on machine learning algorithm can achieve satisfactory prediction efficiency, among which ANNM is the best, which can be used to guide clinicians in diagnosis and treatment and improve the prognosis of EC patients.

SenseHunger: Machine Learning Approach to Hunger Detection Using Wearable Sensors

The perception of hunger and satiety is of great importance to maintaining a healthy body weight and avoiding chronic diseases such as obesity, underweight, or deficiency syndromes due to malnutrition. There are a number of disease patterns, characterized by a chronic loss of this perception. To our best knowledge, hunger and satiety cannot be classified using non-invasive measurements. Aiming to develop an objective classification system, this paper presents a multimodal sensory system using associated signal processing and pattern recognition methods for hunger and satiety detection based on non-invasive monitoring. We used an Empatica E4 smartwatch, a RespiBan wearable device, and JINS MEME smart glasses to capture physiological signals from five healthy normal weight subjects inactively sitting on a chair in a state of hunger and satiety. After pre-processing the signals, we compared different feature extraction approaches, either based on manual feature engineering or deep feature learning. Comparative experiments were carried out to determine the most appropriate sensor channel, device, and classifier to reliably discriminate between hunger and satiety states. Our experiments showed that the most discriminative features come from three specific sensor modalities: Electrodermal Activity (EDA), infrared Thermopile (Tmp), and Blood Volume Pulse (BVP).

Dynamic Pain-Related Changes in Pulse-Graph Measurements in Patients with Primary Dysmenorrhea before and after Electroacupuncture Intervention and Its Correlation with TCM Pattern

OBJECTIVE

To explore the dynamic changes recorded in pulse graph related to the changes in the severity of pain before and after electroacupuncture (EA) intervention among young women suffering from primary dysmenorrhea (PD).

METHODS

A total of 147 female college students were recruited in this study. Based on participants' symptoms associated with menstruation, they were divided into the PD group and the healthy control group. In addition, participants in the PD group were further sorted into the Cold Coagulation and Blood Stasis Pattern (CCBSP) and Qi Stagnation and Blood Stasis Pattern (QSBSP) based on TCM diagnoses and their pulses differences. Participants in the PD group received EA at maximal pain during menstruation. The primary acupuncture points selected were SP 6 and RN 3, additional RN 4 for CCBSP, and LR 3 for QSBSP. Four observation time points were 7-10 days before menstruation (T ₀), maximal pain during menstruation (T ₁), immediately after EA (T ₂), and 30 mins after EA (T ₃). The severity of pain was assessed by a visual analog scale (VAS) along with a pulse analyzer to record the variations of the pulse graph throughout the changes of pain level.

RESULTS

(1) The average VAS score in the PD group decreased from 5.44 ± 1.46 at T ₁ to 1.72 ± 1.27 at T ₂ and 1.59 ± 1.30 at T ₃. The average VAS score in participants of CCBSP at T ₁, T ₂, and T ₃ was higher than that of QSBSP. (2) At T ₁, h ₂, h ₃, h ₄, and w ₁/t were all significantly increased, compared with those at T ₀. At T ₂, t and t ₅ were both significantly increased, and w ₁/t, t ₁, and t ₁/t were all significantly decreased, compared with those at T ₁. At T ₃, w ₁/t, t ₁, and t ₁/t were all significantly increased, and t and t ₅ were both significantly decreased, compared with those at T ₂. (3) Comparing the pulse graphs between the healthy control and the PD groups, h ₁ was significantly lower at T ₀; w ₁/t was significantly higher at T ₁; t was significantly higher at T ₂; and t ₁ and t ₁/t were both significantly higher at T ₃ in PD group. (4) When comparing the pulse graphs between QSBSP and CCBSP, t ₄/t ₅ was significantly higher at T ₀ and t ₁ was significantly higher at T ₁ in the CCBSP group.

CONCLUSION

EA is effective in relieving primary dysmenorrhea. Our results showed the opposite changing of the pulse graph recorded before the onset of pain to the maximum pain and that from maximum pain to pain relief. Indeed, there were differences in the recorded pulse graphs between CCBSP and QSBSP (two patterns of PD) as described in traditional Chinese pulses diagnosis. The study has been registered in the Chinese Clinical Trial Registry (registered number: ChiCTR2000040065; registered date: 2020/11/19).

Evidence-based nursing intervention can improve the treatment compliance, quality of life and self-efficacy of patients with lung cancer undergoing radiotherapy and chemotherapy

OBJECTIVE

To investigate the improvement effect of evidence-based nursing intervention on treatment compliance, quality of life and self-efficacy of patients with lung carcinoma (LC) undergoing radiotherapy and chemotherapy.

METHODS

From May 2018 to August 2019, 183 patients with LC who received radiotherapy and chemotherapy in our hospital were selected and divided into two groups in accordance with different nursing methods. Among them, 85 patients who received routine nursing intervention were included in the control group (CG), and 98 patients who received evidence-based nursing intervention were included in the research group (RG). The improvement of pulmonary function indexes [(FVC), forced expiratory volume in one second (FEV₁), ratio of forced expiratory volume in one second to forced vital capacity (FEV₁/FVC)] was observed before and after nursing. The pain degree was evaluated by the visual analogue scale (VAS). The treatment compliance between groups was compared. The psychological emotions of the patients were evaluated by a self-rating anxiety scale (SAS) and a self-rating depression scale (SDS). The General Self-efficacy Scale (GSES) was applied to assess the self-efficacy and the Quality of Life (SF-36) scale was applied to evaluate the quality of life. The incidence of secondary infection was observed in the two groups. The nursing satisfaction was evaluated by a nursing satisfaction questionnaire made by our hospital.

RESULTS

After nursing, the improvement of FEV₁, FVC and FEV₁/FVC levels in the RG were obviously better than that in the CG; The scores of VAS, SAS and SDS and total incidence of secondary infection in the RG were obviously lower than those in the CG; The treatment compliance, GSES and SF-36 scores, and nursing satisfaction scores of patients in the RG were obviously higher than that in the CG.

CONCLUSION

Evidence-based nursing intervention can improve treatment compliance, lung function, self-efficacy and quality of life for patients with LC undergoing radiotherapy and chemotherapy.

Artificial Neural Networks Model for Predicting Type 2 Diabetes Mellitus Based on VDR Gene FokI Polymorphism, Lipid Profile and Demographic Data

Type 2 diabetes mellitus (T2DM) is a multifactorial disease associated with many genetic polymorphisms; among them is the FokI polymorphism in the vitamin D receptor (VDR) gene. In this case-control study, samples from 82 T2DM patients and 82 healthy controls were examined to investigate the association of the FokI polymorphism and lipid profile with T2DM in the Jordanian population. DNA was extracted from blood and genotyped for the FokI polymorphism by polymerase chain reaction (PCR) and DNA sequencing. Lipid profile and fasting blood sugar were also measured. There were significant differences in high-density lipoprotein (HDL) cholesterol and triglyceride levels between T2DM and control samples. Frequencies of the FokI polymorphism (CC, CT and TT) were determined in T2DM and control samples and were not significantly different. Furthermore, there was no significant association between the FokI polymorphism and T2DM or lipid profile. A feed-forward neural network (FNN) was used as a computational platform to predict the persons with diabetes based on the FokI polymorphism, lipid profile, gender and age. The accuracy of prediction reached 88% when all parameters were included, 81% when the FokI polymorphism was excluded, and 72% when lipids were only included. This is the first study investigating the association of the VDR gene FokI polymorphism with T2DM in the Jordanian population, and it showed negative association. Diabetes was predicted with high accuracy based on medical data using an FNN. This highlights the great value of incorporating neural network tools into large medical databases and the ability to predict patient susceptibility to diabetes.

Plasma acyl ghrelin and nonesterified fatty acids are the best predictors for hunger status in pregnant gilts

Sows are usually restricted fed during pregnancy to maximize their reproductive efficiency, which may predispose sows to a state of hunger. However, an objective measurement of hunger status has not been established. In the present study, we examined the correlation of plasma hormones and NEFA and selected the best predictors for hunger status using pregnant gilts. Three different levels of feed intake (0.5, 1.0 and 2.0 × maintenance energy intake [0.5M, 1.0M and 2.0M, respectively]) were imposed from Day 28 to 34 of gestation to create different hunger statuses in pregnant gilts. Plasma hormones related to energy homeostasis and NEFA were analyzed to quantify their response to different levels of feed intake. A total of 18 gilts (197.53 ± 6.41 kg) were allotted to 1 of 3 dietary treatments using a completely randomized design. Results showed that BW change, ADG, and G:F from Day 28 to 34 of gestation were higher ( < 0.01) for gilts on the 2.0M feeding level than for gilts on the 0.5M feeding level. Plasma acyl ghrelin concentrations showed a relatively flat pattern during the 24-h period. Plasma acyl ghrelin and NEFA concentrations and areas under the curve (AUC) were greater ( < 0.05) in gilts on the 0.5M level of feed intake than in those on the 2.0M level of feed intake. No differences were observed among the 3 feeding levels in terms of plasma glucagon-like peptide 1 and leptin concentrations. Additionally, consumption time for 1.82 kg feed on Day 35 of gestation was longer ( < 0.01) in gilts fed the 2.0M level of feed intake from Day 28 to 34 of gestation than in those on the 0.5M level of feed intake. Simple linear regression results showed that the AUC of acyl ghrelin was the best predictor for consumption time ( = 0.82), whereas the AUC of NEFA was the best predictor for BW ( = 0.55) or backfat change ( = 0.42) from Day 28 to 34 of gestation. In conclusion, our data suggested that a relative flat pattern existed in pregnant gilts in terms of the diurnal plasma profile of acyl ghrelin and that the level of feed intake of pregnant gilts was negatively correlated with plasma concentrations of acyl ghrelin and NEFA, which, in turn, were negatively associated with feed consumption time. The AUC of acyl ghrelin and NEFA seemed to be the best predictors for hunger status of pregnant gilts.