Search Results (38)

Vending machines have evolved into a critical element of the intelligent healthcare service system. To enhance the precision of pill box detection counting and cater to the lightweight requirements of its internal embedded controller for deep learning frameworks, an enhanced lightweight YOLOv8n model is introduced. A dataset comprising 4080 images is initially compiled for model training and assessment purposes. The refined YOLOv8n-ShuffleNetV2 model is crafted, featuring the integration of ShuffleNetv2 as the new backbone network, the incorporation of the VoVGSCSP module to bolster feature extraction capabilities, and the utilization of the Wise-IoU v3 loss function for bounding box regression enhancement. Moreover, a model pruning strategy based on structured pruning (SFP) and layer-wise adaptive magnitude pruning (LAMP) is implemented. Comparative experimental findings demonstrate that the enhanced and pruned model has elevated the mean Average Precision (mAP) rate from 94.5% to 95.1%. Furthermore, the model size has been reduced from 11.1 MB to 6.0 MB, and the inference time has been notably decreased from 1.97 s to 0.34 s. The model’s accuracy and efficacy are validated through experiments conducted on the Raspberry Pi 4B platform. The outcomes of the experiments underscore how the refined model significantly amplifies the deployment efficiency of the deep learning model on resource-limited devices, thus greatly supporting the advancement of intelligent medicine management and medical vending machine applications. Full article

The accurate identification of pills is essential for their safe administration in the medical field. Despite technological advancements, pill classification encounters hurdles such as ambiguous images, pattern similarities, mixed pills, and variations in pill shapes. A significant factor is the inability of 2D imaging to capture a pill’s 3D structure efficiently. Additionally, the scarcity of diverse datasets reflecting various pill shapes and colors hampers accurate prediction. Our experimental investigation shows that while color-based classification obtains a 95% accuracy rate, shape-based classification only reaches 66%, underscoring the inherent difficulty distinguishing between pills with similar patterns. In response to these challenges, we propose a novel system integrating Multi Combination Pattern Labeling (MCPL), a new method designed to accurately extract feature points and pill patterns. MCPL extracts feature points invariant to rotation and scale and effectively identifies unique edges, thereby emphasizing pills’ contour and structural features. This innovative approach enables the robust extraction of information regarding various shapes, sizes, and complex pill patterns, considering even the 3D structure of the pills. Experimental results show that the proposed method improves the existing recognition performance by about 1.2 times. By improving the accuracy and reliability of pill classification and recognition, MCPL can significantly enhance patient safety and medical efficiency. By overcoming the limitations inherent in existing classification methods, MCPL provides high-accuracy pill classification, even with constrained datasets. It substantially enhances the reliability of pill classification and recognition, contributing to improved patient safety and medical efficiency. Full article

Background: Wireless capsule endoscopy (WCE) has significantly advanced the diagnosis of gastrointestinal (GI) diseases by allowing for the non-invasive visualization of the entire small intestine. However, machine learning-based methods for organ classification in WCE often rely on color information, leading to decreased performance when obstacles such as food debris are present. This study proposes a novel model that integrates convolutional neural networks (CNNs) and long short-term memory (LSTM) networks to analyze multiple frames and incorporate temporal information, ensuring that it performs well even when visual information is limited. Methods: We collected data from 126 patients using PillCam™ SB3 (Medtronic, Minneapolis, MN, USA), which comprised 2,395,932 images. Our deep learning model was trained to identify organs (stomach, small intestine, and colon) using data from 44 training and 10 validation cases. We applied calibration using a Gaussian filter to enhance the accuracy of detecting organ boundaries. Additionally, we estimated the transit time of the capsule in the gastric and small intestine regions using a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) designed to be aware of the sequence information of continuous videos. Finally, we evaluated the model’s performance using WCE videos from 72 patients. Results: Our model demonstrated high performance in organ classification, achieving an accuracy, sensitivity, and specificity of over 95% for each organ (stomach, small intestine, and colon), with an overall accuracy and F1-score of 97.1%. The Matthews Correlation Coefficient (MCC) and Geometric Mean (G-mean) were used to evaluate the model’s performance on imbalanced datasets, achieving MCC values of 0.93 for the stomach, 0.91 for the small intestine, and 0.94 for the colon, and G-mean values of 0.96 for the stomach, 0.95 for the small intestine, and 0.97 for the colon. Regarding the estimation of gastric and small intestine transit times, the mean time differences between the model predictions and ground truth were 4.3 ± 9.7 min for the stomach and 24.7 ± 33.8 min for the small intestine. Notably, the model’s predictions for gastric transit times were within 15 min of the ground truth for 95.8% of the test dataset (69 out of 72 cases). The proposed model shows overall superior performance compared to a model using only CNN. Conclusions: The combination of CNN and LSTM proves to be both accurate and clinically effective for organ classification and transit time estimation in WCE. Our model’s ability to integrate temporal information allows it to maintain high performance even in challenging conditions where color information alone is insufficient. Including MCC and G-mean metrics further validates the robustness of our approach in handling imbalanced datasets. These findings suggest that the proposed method can significantly improve the diagnostic accuracy and efficiency of WCE, making it a valuable tool in clinical practice for diagnosing and managing GI diseases. Full article

The aim of this study was to assess whether dietary supplementation with a nutraceutical blend comprising extracts of bergamot and artichoke—both standardized in their characteristic polyphenolic fractions—could positively affect serum lipid concentration and insulin sensitivity, high-sensitivity C-reactive protein (hs-CRP), and indexes of non-alcoholic fatty liver disease (NAFLD) in 90 healthy individuals with suboptimal cholesterol levels. Participants were randomly allocated to treatment with a pill of either active treatment or placebo. After 6 weeks, the active-treated group experienced significant improvements in levels of triglycerides (TG), apolipoprotein B-100 (Apo B-100), and apolipoprotein AI (Apo AI) versus baseline. Total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), non-high density lipoprotein cholesterol (Non-HDL-C), and hs-CRP also significantly decreased in the active-treated group compared to both baseline and placebo. At the 12-week follow-up, individuals allocated to the combined nutraceutical experienced a significant improvement in TC, LDL-C, Non-HDL-C, TG, Apo B-100, Apo AI, glucose, alanine transaminase (ALT), gamma-glutamyl transferase (gGT), hs-CRP, several indexes of NAFLD, and brachial pulse volume (PV) in comparison with baseline. Improvements in TC, LDL-C, Non-HDL-C, TG, fatty liver index (FLI), hs-CRP, and endothelial reactivity were also detected compared to placebo (p < 0.05 for all). Overall, these findings support the use of the tested dietary supplement containing dry extracts of bergamot and artichoke as a safe and effective approach for the prevention and management of a broad spectrum of cardiometabolic disorders. Full article

Background and Objectives: The incidence of diabetic osteoporosis, an important complication of diabetes mellitus, is increasing gradually. This study investigated the combined effect of the Zuogui pill (ZGP) and eldecalcitol (ED-71), a novel vitamin D analog, on type 2 diabetic osteoporosis (T2DOP) and explored their action mechanism. Materials and Methods: Blood glucose levels were routinely monitored in db/db mice while inducing T2DOP. We used hematoxylin and eosin staining, Masson staining, micro-computed tomography, and serum biochemical analysis to evaluate changes in the bone mass and blood calcium and phosphate levels of mice. Immunohistochemical staining was performed to assess the osteoblast and osteoclast statuses. The MC3T3-E1 cell line was cultured in vitro under a high glucose concentration and induced to undergo osteogenic differentiation. Quantitative real-time polymerase chain reaction, Western blot, immunofluorescence, ALP, and alizarin red staining were carried out to detect osteogenic differentiation and PI3K–AKT signaling pathway activity. Results: ZGP and ED-71 led to a dramatic decrease in blood glucose levels and an increase in bone mass in the db/db mice. The effect was strongest when both were used together. ZGP combined with ED-71 promoted osteoblast activity and inhibited osteoclast activity in the trabecular bone region. The in vitro results revealed that ZGP and ED-71 synergistically promoted osteogenic differentiation and activated the PI3K–AKT signaling pathway. The PI3K inhibitor LY294002 or AKT inhibitor ARQ092 altered the synergistic action of both on osteogenic differentiation. Conclusions: The combined use of ZGP and ED-71 reduced blood glucose levels in diabetic mice and promoted osteogenic differentiation through the PI3K–AKT signaling pathway, resulting in improved bone mass. Our study suggests that the abovementioned combination constitutes an effective treatment for T2DOP. Full article

The aim of this narrative review is to summarize the current pharmacotherapeutic treatment options for osteoarthritis (OA). Is therapy still mainly symptomatic or does the pill against arthrosis already exist? Causal and non-causal, as well as future therapeutic approaches, are discussed. Various surgical and non-surgical treatment options are available that can help manage symptoms, slow down progression, and improve quality of life. To date, however, therapy is still mainly symptomatic, often using painkilling and anti-inflammatory drugs until the final stage, which is usually joint replacement. These “symptomatic pills against” have side effects and do not alter the progression of OA, which is caused by an imbalance between degenerative and regenerative processes. Next to resolving mechanical issues, the goal must be to gain a better understanding of the cellular and molecular basis of OA. Recently, there has been a lot of interest in cartilage-regenerative medicine and in the current style of treating rheumatoid arthritis, where drug therapy (“the pill against”) has been established to slow down or even stop the progression of rheumatoid arthritis and has banned the vast majority of former almost regular severe joint destructions. However, the “causal pill against” OA does not exist so far. First, the early detection of osteoarthritis by means of biomarkers and imaging should therefore gain more focus. Second, future therapeutic approaches have to identify innovative therapeutic approaches influencing inflammatory and metabolic processes. Several pharmacologic, genetic, and even epigenetic attempts are promising, but none have clinically improved causal therapy so far, unfortunately. Full article

Crohn’s disease (CD) is a chronic inflammatory disorder characterized by a transmural inflammation that may involve any part of the gastrointestinal tract. An evaluation of small bowel involvement, allowing recognition of disease extent and severity, is important for disease management. Current guidelines recommend the use of capsule endoscopy (CE) as a first-line diagnosis method for suspected small bowel CD. CE has an essential role in monitoring disease activity in established CD patients, as it can assess response to treatment and identify high-risk patients for disease exacerbation and post-operative relapse. Moreover, several studies have shown that CE is the best tool to assess mucosal healing as part of the treat-to-target strategy in CD patients. The PillCam Crohn’s capsule is a novel pan-enteric capsule which enables visualization of the whole gastrointestinal tract. It is useful to monitor pan-enteric disease activity, mucosal healing and accordingly allows for the prediction of relapse and response using a single procedure. In addition, the integration of artificial intelligence algorithms has showed improved accuracy rates for automatic ulcer detection and the ability to shorten reading times. In this review, we summarize the main indications and virtue for using CE for the evaluation of CD, as well as its implementation in clinical practice. Full article

Medication should be consumed as prescribed with little to zero margins for errors, otherwise consequences could be fatal. Due to the pervasiveness of camera-equipped mobile devices, patients and practitioners can easily take photos of unidentified pills to avert erroneous prescriptions or consumption. This area of research goes under the umbrella of information retrieval and, more specifically, image retrieval or recognition. Several studies have been conducted in the area of image retrieval in order to propose accurate models, i.e., accurately matching an input image with stored ones. Recently, neural networks have been shown to be effective in identifying digital images. This study aims to provide an enhancement to image retrieval in terms of accuracy and efficiency through image segmentation and classification. This paper suggests three neural network (CNN) architectures: two models that are hybrid networks paired with a classification method (CNN+SVM and CNN+kNN) and one ResNet-50 network. We perform various preprocessing steps by using several detection techniques on the selected dataset. We conduct extensive experiments using a real-life dataset obtained from the National Library of Medicine database. The results demonstrate that our proposed model is capable of deriving an accuracy of 90.8%. We also provide a comparison of the above-mentioned three models with some existing methods, and we notice that our proposed CNN+kNN architecture improved the pill image retrieval accuracy by 10% compared to existing models. Full article

Rapid urban development poses a threat to global biodiversity. At the same time, urban green spaces offer opportunities for holding biodiversity in cities. Among biological communities, the soil fauna plays a crucial role in ecological processes but is often ignored. Understanding the effects of environmental factors on soil fauna is critical for ecological conservation in urban areas. In this study, five typical green space habitats were selected including bamboo grove, forest, garden, grassland, and wasteland in spring, for detecting the relationship between habitats and Armadillidium vulgare population characteristics in Yancheng, China. Results indicate that soil water content, pH, soil organic matter, and soil total carbon varied significantly among habitats, as well as the body length and body weight of pill bugs. The higher proportion of larger pill bugs was found in the wasteland and the lower proportion in the grassland and the bamboo grove. The body length of pill bugs was positively related to pH. Soil total carbon, soil organic matter, and the number of plant species were correlated with the body weight of pill bugs. Full article

The recognition process of natural scenes is complicated at present, and images themselves may be complex owing to the special features of natural scenes. In this study, we use the detection and recognition of pill box text as an application scenario and design a deep-learning-based text detection algorithm for such natural scenes. We propose an end-to-end graphical text detection and recognition model and implement a detection system based on the B/S research application for pill box recognition, which uses DBNet as the text detection framework and a convolutional recurrent neural network (CRNN) as the text recognition framework. No prior image preprocessing is required in the detection and recognition processes. The recognition result from the back-end is returned to the front-end display. Compared with traditional methods, this recognition process reduces the complexity of preprocessing prior to image detection and improves the simplicity of the model application. Experiments on the detection and recognition of 100 pill boxes demonstrate that the proposed method achieves better accuracy in text localization and recognition results than the previous CTPN + CRNN method. The proposed method is significantly more accurate and easier to use than the traditional approach in terms of both training and recognition processes. Full article

A solid-contact potentiometric sensor has been developed on the basis of glassy carbon electrode covered with electropolymerized polyaniline and alternatively layered pillar[6]arene and Cu²⁺ ions films. The assembly of the surface layer was confirmed by surface plasmon resonance measurements. The number of deposited layers was selected to reach better analytical characteristics for Cu²⁺ determination. It was shown that better results were achieved by using five layers, the upper one consisting of the macrocycle. The addition of covering layers for polyelectrolytes (Nafion, poly(styrene sulfonate)) and Cu²⁺ ions did not improve sensor performance. The potentiometric sensor made it possible to determine Cu²⁺ ions in neutral and weakly acidic media with a linear range of the concentrations, from 3.0 μM to 10.0 mM (limit of detection 3.0 μM). The applicability of the sensor in real sample assays was confirmed by the determination of Cu²⁺ ions in copper vitriol, Bordeaux mixture, and polyvitamin-mineral pills of “Complivit” during an atomic emission spectroscopy analysis. Full article

Gandouling (GDL) Pill is a novel Traditional Chinese medicinal drug to treat Wilson’s disease in clinics. It is composed of six separate herbal medicines, including Rhei Radix ET Rhizoma, Coptidis Rhizoma, Salviae Miltiorrhizae Radix ET Rhizoma, Spatholobi Caulis, Curcumae Rhizoma, and Curcumae Longae Rhizoma. In this study, a strategy was proposed to investigate the chemical constituents and to quantify the potential bioactive components in GDL Pill. Firstly, the mass fragmentation behaviors of representative compounds were investigated, and, in total, 69 compounds were characterized in GDL Pill using full scan/dd-MS² scan mode by ultra-high-performance liquid chromatography (UPLC)/Q-Orbitrap mass spectrometry (MS). These compounds included 18 alkaloids, 18 ketones, 16 phenolic compounds, 11 organic acids, and 6 tanshinones. Seventeen of the compounds were unambiguously identified by comparison with reference standards. Secondly, the absorption components of GDL Pill in rat plasma were investigated by using target-Selected Ion Monitoring (t-SIM) scan mode built in Q-Orbitrap MS. A total of 18 components were detected, which were considered as potential bioactive components of GDL Pill. Thirdly, 10 major absorption components were simultaneously determined in six batches of samples by UPLC/diode array detector (DAD). The method was fully validated with respect to linearity, precision, repeatability, stability, and recovery. Alkaloids from Coptidis Rhizoma, such as coptisine (8), berberine (18), palmatine (19), were the most abundant bioactive compounds for GDL Pill that possess the potential be used as quality markers. The proposed strategy is practical and efficient for revealing the material basis of GDL Pill, and also provides a simple and accurate method for quality control. Full article

Object counting is an indispensable task in manufacturing and management. Recently, the development of image-processing techniques and deep learning object detection has achieved excellent performance in object-counting tasks. Accordingly, we propose a novel small-size smart counting system composed of a low-cost hardware device and a cloud-based object-counting software server to implement an accurate counting function and overcome the trade-off presented by the computing power of local hardware. The cloud-based object-counting software consists of a model adapted to the object-counting task through a novel DBC-NMS (our own technique) and hyperparameter tuning of deep-learning-based object-detection methods. With the power of DBC-NMS and hyperparameter tuning, the performance of the cloud-based object-counting software is competitive over commonly used public datasets (CARPK and SKU110K) and our custom dataset of small pills. Our cloud-based object-counting software achieves an mean absolute error (MAE) of 1.03 and a root mean squared error (RMSE) of 1.20 on the Pill dataset. These results demonstrate that the proposed smart counting system accurately detects and counts densely distributed object scenes. In addition, the proposed system shows a reasonable and efficient cost–performance ratio by converging low-cost hardware and cloud-based software. Full article

Patients usually deviate from prescribed medication schedules and show reduced adherence. Even when the adherence is sufficient, there are conditions where the medication schedule should be modified. Crucial drug–drug, food–drug, and supplement–drug interactions can lead to treatment failure. We present the development of an internet of medical things (IoMT) platform to improve medication adherence and enable remote treatment modifications. Based on photos of food and supplements provided by the patient, using a camera integrated to a portable 3D-printed low-power pillbox, dangerous interactions with treatment medicines can be detected and prevented. We compare the medication adherence of 14 participants following a complex medication schedule using a functional prototype that automatically receives remote adjustments, to a dummy pillbox where the adjustments are sent with text messages. The system usability scale (SUS) score was 86.79, which denotes excellent user acceptance. Total errors (wrong/no pill) between the functional prototype and the dummy pillbox did not demonstrate any statistically significant difference (p = 0.57), but the total delay of the intake time was higher (p = 0.03) during dummy pillbox use. Thus, the proposed low-cost IoMT pillbox improves medication adherence even with a complex regimen while supporting remote dose adjustment. Full article

Sulfonylureas, a family of anti-diabetic drugs widely used in the clinical treatment of type 2 diabetes, have recently emerged as an illegal adulterant in functional foods, to enhance the claimed anti-diabetic activity. To establish a screening assay method against their adulteration, with the aid of molecular simulation of hapten, two antibodies were raised and complementarily used to enhance the broad-specificity of an enzyme-linked immunosorbent assay (ELISA), which demonstrated simultaneous detection capability to 6 sulfonylureas; the detection limits ranged from 0.02 to 1.0 ng/mL, and recoveries were between 78.3% to 104.5%. Liquid chromatography with tandem mass spectrometry (LC-MS/MS) confirmed the reliability of the proposed ELISA, based on real samples. These results suggest that the proposed ELISA could be an ideal method for screening to monitor for illicit adulteration of sulfonylureas in functional pill products. Full article