Low Accuracy Research Articles

Solutions to elliptic and parabolic problems via finite difference based unsupervised small linear convolutional neural networks

In recent years, there has been a growing interest in leveraging deep learning and neural networks to address scientific problems, particularly in solving partial differential equations (PDEs). However, many neural network-based methods like PINNs rely on auto differentiation and sampling collocation points, leading to a lack of interpretability and lower accuracy than traditional numerical methods. As a result, we propose a fully unsupervised approach, requiring no training data, to estimate finite difference solutions for PDEs directly via small linear convolutional neural networks. Our proposed approach uses substantially fewer parameters than similar finite difference-based approaches while also demonstrating comparable accuracy to the true solution for several selected elliptic and parabolic problems compared to the finite difference method.

Forensic bone age assessment of hand and wrist joint MRI images in Chinese han male adolescents based on deep convolutional neural networks.

In Chinese criminal law, the ages of 12, 14, 16, and 18 years old play a significant role in the determination of criminal responsibility. In this study, we developed an epiphyseal grading system based on magnetic resonance image (MRI) of the hand and wrist for the Chinese Han population and explored the feasibility of employing deep learning techniques for bone age assessment based on MRI of the hand and wrist. This study selected 282 Chinese Han Chinese males aged 6.0-21.0 years old. In the course of our study, we proposed a novel deep learning model for extracting and enhancing MRI hand and wrist bone features to enhance the prediction of target MRI hand and wrist bone age and achieve precise classification of the target MRI and regression of bone age. The evaluation metric for the classification model including precision, specificity, sensitivity, and accuracy, while the evaluation metrics chosen for the regression model are MAE. The epiphyseal grading was used as a supervised method, which effectively solved the problem of unbalanced sample distribution, and the two experts showed strong consistency in the epiphyseal plate grading process. In the classification results, the accuracy in distinguishing between adults and minors was 91.1%, and the lowest accuracy in the three minor classifications (12, 14, and 16 years of age) was 94.6%, 91.1% and 96.4%, respectively. The MAE of the regression results was 1.24 years. In conclusion, the deep learning model proposed enabled the age assessment of hand and wrist bones based on MRI.

Pasteables: A Flexible and Smart “Stick-and-Peel” Wearable Platform for Fitness and Athletics

Wearable technologies are becoming increasingly popular and transforming our fitness and sports industry. Monitoring health parameters and body activity can help achieve fitness goals, improve sports performance, and aid in physiotherapy. However, state-of-the-art wearable devices are often not flexible, expensive, hard to set up, or have privacy concerns. They are designed and optimized for a specific application with tight hardware-software integration and it is hard or impossible to re-purpose them for diverse use cases. In this paper, we propose a flexible, re-configurable human body movement and health monitoring platform, called "Pasteables". It is a stick-and-peel device (which acts like a band-aid) that attaches to the skin or clothing and can create an on-body network of smart wearables. Given an application, Pasteables can be easily adapted to its requirements while ensuring good performance and privacy. We present the overall architecture, system design steps, and the configuration process. We have designed a set of functional prototypes and configured them for performing a variety of camera-less pose and posture detection tasks, which are important for athletes, professional dancing, physiotherapy, and fitness workouts. We note that the Pasteables can perform pose/posture detection without external stationary reference at low cost and high accuracy.

Recognition of Heat-Damaged Corn Seeds Based on Fusion of Laser Ultrasonic Signal and Infrared Image Features

Corn is widely cultivated on a global scale. However, high temperatures during storage and transportation can lead to thermal damage to the kernels, negatively impacting their quality. Traditional methods for identifying heat-damaged grains primarily rely on manual inspection, which is characterized by low efficiency and accuracy. This study proposes a novel identification method that integrates laser ultrasonic signals with infrared image texture features. A pulsed laser stimulates the seeds to generate laser ultrasonic signals, while an infrared camera captures infrared images of the seeds. We extract time-domain, frequency-domain, and Hilbert-domain features from the laser ultrasonic signals, in addition to texture features from the infrared images. These features are combined using Canonical Correlation Analysis (CCA). Subsequently, the fused features are classified using a Backpropagation (BP) neural network, Support Vector Machine (SVM), and Particle Swarm Optimization–Support Vector Machine (PSO–SVM). The results indicate that the recognition rate achieved with the fused ‘signal-image’ features reaches 99.17%, providing a novel approach for detecting heat-damaged corn seeds.

Lithium battery SOC estimation based on improved sparrow search algorithm and backpropagation neural network

Accurate state-of-charge (SOC) estimation is crucial for optimal battery management. This paper proposes a novel method, the Improved Sparrow Search Algorithm-Backpropagation (ISSA-BP) neural network, to address the issue of low estimation accuracy encountered with a single BP neural network. ISSA is used to optimize the initial weights and thresholds of the BP neural network, effectively overcoming its tendency to get stuck in local minima. Compared to the single BP neural network, ISSA-BP demonstrates significantly improved accuracy under two conditions (DST and BJDST), with reductions in root mean square error by 64.0% and 50.9% and mean absolute error by 69.8% and 51.1%, respectively. These results highlight the superior robustness and accuracy of the ISSA-BP algorithm for SOC estimation in lithium batteries.

Data-driven parallel optimization method for sensor placement and state estimation based on generative adversarial principle in unobservable systems

Abstract Data-Driven State Estimation (DDSE) is characterized by low measurement requirements and high estimation accuracy, which provides the possibility of stable and efficient estimation for unobservable distribution systems with low measurement quantities. However, the lack of an efficient and flexible sensor placement method with DDSE at this stage hinders the development and application of DDSE in distribution networks. To address this issue, a data-driven parallel optimization method for sensor placement and state estimation based on the generative adversarial principle in unobservable systems is proposed in this paper. This method transforms sensor placement and state estimation into an adversarial learning model, solvable through gradient descent. The sensor placement optimization strategy generates a cost-effective placement scheme, while DDSE evaluates the sensor scheme, provides improvement suggestions, and enhances its estimation accuracy. Through adversarial training, the proposed scheme yields a significantly reduced number of sensor placement schemes compared to traditional methods, while maintaining high precision of DDSE. Simulation results on the IEEE-33 node system validate the effectiveness of the proposed approach.

Noninvasive, epigenetic age estimation in an elasmobranch, the cownose ray (Rhinoptera bonasus)

Age data are essential for estimating life history parameters and are thus critical for population assessment, management, and conservation. Traditional vertebrae-based age estimation in elasmobranchs can be costly, time intensive, of low accuracy, and is by necessity lethal. Herein, epigenetic clocks were developed for an elasmobranch, the cownose ray (Rhinoptera bonasus), using aquarium-born individuals (n = 42) with known dates of birth (age range: 7−7,878 days or 0−21 years) and two tissue types (fin clips and whole blood) that can be sampled in a relatively non-invasive manner. Enzymatically-converted restriction site-associated DNA sequencing (ECrad-seq) was used to identify CpG sites that exhibited age-correlated DNA methylation. The epigenetic clocks developed were highly accurate (mean absolute error, MAE, < 0.75 years) and precise (R2 > 0.98). Age-associated CpG sites were identified across tissues, and a multi-tissue clock was also highly accurate (MAE < 1 year) and precise (R2 = 0.97). Using the developed fin clip clock, three wild-caught individuals of unknown age but managed in aquariums for > 22 years were predicted to be 22.10−23.49 years old. Overall, the results have important implications for future epigenetic clock development and noninvasive age estimation in elasmobranchs.

Ready-to-use Models Built Using a Diverse Set of 266 Aroma Compounds for the Estimation of Gas Chromatographic Retention Indices for the 50%-Cyanopropylphenyl-50%-Dimethylpolysiloxane Stationary Phase.

Retention index prediction based on the molecule structure is not often used in practice due to low accuracy, the need to use paid software to calculate molecular descriptors (MD), and the narrow applicability domain of many models. In recent years, relatively accurate and versatile deep learning (DL)-based models have emerged. These models are now used in practice as an additional criterion in gas chromatography-mass spectrometry identification. The DB-225ms stationary phase (usually described as 50%-cyanopropylphenyl-50%-dimethylpolysiloxane in available sources) is widely used, but ready-to-use retention index estimation models are not available for it. This study presents such models. The models are linear and use simple constitutional MD and retention indices predicted by DL for the DB-WAX and DB-624 stationary phases as MD (we show that it is their use that allows us to achieve satisfactory accuracy). The accuracy obtained for a completely unseen hold-out test set: root mean square error 73.2; mean absolute error 45.7; median absolute error 22.0. The models were trained using a retention data set of 266 volatile compounds. All calculations can be performed using the convenient open-source software CHERESHNYA. The final equations are implemented as a spreadsheet and a code snippet and are available online: https://doi.org/10.6084/m9.figshare.26800789.

Is ChatGPT a reliable tool in Autoimmune Hepatitis?

Artificial intelligence-based chatbots offer a potential avenue for delivering personalized counselling to Autoimmune Hepatitis (AIH) patients. We assessed accuracy, completeness, comprehensiveness and safety of ChatGPT-4 responses to 12 inquiries out of a pool of 40 questions posed by four AIH patients. Questions were categorized into three areas: Diagnosis(1-3), Quality of Life(4-8) and Medical treatment(9-12). 11 Key Opinion Leaders (KOLs) evaluated responses using a Likert scale with 6 points for accuracy, 5 points for safety and 3 points for completeness and comprehensiveness. Median scores for accuracy, completeness, comprehensiveness and safety were 5(4-6), 2 (2-2) and 3 (2-3); no domain exhibited superior evaluation. Post-diagnosis follow-up question was the trickiest with low accuracy and completeness but safe and comprehensive features. Agreement among KOLs (Fleiss's Kappa statistics) was slight for accuracy (0.05) but poor for the remaining features (-0.05, -0.06 and -0,02, respectively). Chatbots show good comprehensibility but lack reliability. Further studies are needed to integrate Chat-GPT within clinical practice.

The performance of ChatGPT-4.0o in medical imaging evaluation: a preliminary investigation

This study investigated the performance of ChatGPT-4.0o in evaluating the quality of positioning in radiographic images. Thirty radiographs depicting a variety of knee, elbow, ankle, hand, pelvis, and shoulder projections were produced using anthropomorphic phantoms and uploaded to ChatGPT-4.0o. The model was prompted to provide a solution to identify any positioning errors with justification and offer improvements. A panel of radiographers assessed the solutions for radiographic quality based on established positioning criteria, with a grading scale of 1-5. In only 20% of projections, ChatGPT-4.0o correctly recognized all errors with justifications and offered correct suggestions for improvement. The most commonly occurring score was 3 (9 cases, 30%), wherein the model recognized at least 1 specific error and provided a correct improvement. The mean score was 2.9. Overall, low accuracy was demonstrated, with most projections receiving only partially correct solutions. The findings reinforce the importance of robust radiography education and clinical experience.

Evaluation of Self-vision Assessment Charts in Schools of Eastern Nepal: A Multi Method Study

Introduction: Globally, over two billion people suffer from vision impairment, almost half preventable. In Nepal, the Eastern Regional Eye Care Program introduced self-vision assessment charts in 137 schools for early eye issue detection. This study assessed the charts' use and perceived impact.Methods: A multi-method study was conducted to evaluate the use of self-vision assessment charts in schools in Eastern Nepal. Ethical approval was obtained from Institutional Review Board (Refrence number: 72-079-040). Quantitative data were collected through a questionnaire survey. Quantitative data were analyzed using descriptive statistics, while qualitative data were analyzed using framework analysis. Both methods were integrated to enhance understanding of participants' experiences.Results: In this study, 400 students participated in questionnaire survey, 180 in self vision assessment, and 16 teachers in in-depth interviews. Among the participants, 198 (49.50%) of students used the chart, 125 (65.78%) expressed satisfaction, 241 (60.25%) found it useful, 276 (69%) as beneficial, and 253 (63.25%) stated improved health. There were 93 (23.25%) students who felt knowledgeable about its use, with 18 (10%) using it correctly. In 162 (90%) of schools there were adequate chart placement, 23 (12.77%) provided clear signage for viewing points. Teachers recognized the chart's benefits and encouraged its use, but students lacked awareness. Regular use was viewed as beneficial for early detection of eye issues, though challenges included a need for more awareness among students and teachers and reliance on home remedies.Conclusions: Most students and teachers acknowledged self-vision assessment charts’ benefits but had low utilization and accuracy rates. Strategic interventions such as regular informational sessions, teacher motivation, and engagement with eye health experts are essential.

Accuracy of repaired maxillary dentures from two different repairing techniques: In vitro study

BackgroundDenture fracture is a common problem with acrylic dentures. The fractured denture can be repaired using various techniques such as self-cure acrylic resin acrylic resin and fiber-reinforced acrylic resin. PurposeThe purpose of this study was to compare the accuracy of dentures repaired with self-cure acrylic resin and fiber-reinforced acrylic resin processed using two different techniques (long-cure and microwave processing). Materials and methodsA total of 20 maxillary complete dentures were processed with two techniques; heat (long cycle) processing (10 dentures) and microwave processing (10 dentures). The maxillary cast and denture surface were scanned with Medit intraoral (Medit i700, Medit, South Korea) and STL files were created. Then, the dentures were sectioned at the midline and repaired using self-cure acrylic resin and fiber-reinforced acrylic resin and scanned with Medit intraoral. Finally, adaptation deviations were analyzed from computer software (Geomagic Control X, 3D Systems Inc., USA). The adaptation deviations in each group (long cure and microwave) were compared using an Independent T-test. Two-way ANOVA was done to see whether curing techniques and repairing methods affect the accuracy of repair. A P-value of 0.05 was considered significant. ResultsThe adaptation deviation was slightly higher in the fiber-reinforced acrylic resin group (0.565 ± 0.093) than in the self-cure acrylic resin group (0.536 ± 0.066). However, there was no statistical difference in the adaptation deviations of repaired dentures with self-cure acrylic resin and fiber-reinforced acrylic resin in the long-curing (P-value 0.245) and the microwave (P-value 0.638). Similarly, the adaptation deviation was slightly higher in the long-curing group (0.577 ± 0.075) than in the microwave group (0.524 ± 0.079). However, there was a statistically significant difference in the adaptation deviation of repaired dentures between long-curing and microwave techniques with self-cure acrylic resin (P-value 0.016) but no difference in fiber-reinforced acrylic resin (P-value 0.127). The result of Two-way ANOVA shows that there is no statistically significant interaction between curing techniques (long curing and microwave) and repairing methods (self-cure acrylic resin and fiber-reinforced acrylic resin) for adaptation deviations (P-value 0.646). However, the curing techniques show statistically significant differences (P-value 0.039). ConclusionAcrylic dentures can be repaired with self-cure resin or fiber-reinforced self-cure resin using various processing methods. The accuracy of the denture after repair is unaffected by the repairing method (self-cure acrylic resin and fiber-reinforced acrylic resin) but the accuracy of the denture after repair is affected by the curing techniques (long-curing and microwave). In self-cure resin, the microwave processing showed higher adaptation deviation and less accuracy, whereas the long-curing processing showed lower adaptation deviation and high accuracy.

Harnessing Machine Learning for Effective Waste Classification and Recycling

The world produces nearly 3.5 million tons waste every day so it is important to improve waste management and recycling system. This paper defines how machine learning can classify waste, specifically using VGG16 CNN model to classify 9 types of recyclable materials such as light bulbs, paper, plastic, organic, glass, batteries, clothes, metal and e-waste .We had used nearly 8000+ images from Google Images and Dreamstime.com to test VGG16 model and compare it with another model named as Inceptionv3. But Inceptionv3 showed higher accuracy (~90%) ,and it did not work well when tested on real data. In contrast, the VGG16 model which initially had lower accuracy of (~70%), performed better when tested on real-world scenarios. We also developed a web application which not only classifies the waste but also provides the valuable information to help people how to minimize waste reduction and maximize recycling. The classified images of the waste and its respective findings will also be shared with local authorities to enhance better recycling efforts. Hence this study focuses on need to select right models for waste classification and suggests looking into another machine learning techniques in future.

The benefits and accuracy of real-time continuous glucose monitoring in children and adolescents with type 1 diabetes attending a summer camp.

This study evaluated the usability, satisfaction, and accuracy of a real-time continuous glucose monitoring (rt-CGM) in children and adolescents with type 1 diabetes (T1D) attending a summer camp. Seven children and adolescents with T1D (camper) and 31 of healthcare providers (HCPs) participating in a 2-day summer camp in Kumamoto, Japan were enrolled. The usability and satisfaction were evaluated by tailored questionnaire. The accuracy of rt-CGM was evaluated using self-monitoring of blood glucose (BG) and sensor glucose (SG) values before or after (off camp) and during (on camp) the camp. The score of the usefulness of rt-CGM showed 3.29 ± 0.90 in campers and 4.23 ± 0.87 in HCPs (P = 0.017). The degree of recommendation score for rt-CGM was 3.29 ± 1.11 in campers and 4.23 ± 0.79 in HCPs (P = 0.013). Time in range (TIR) off camp was 45.9% and that on camp was 57.0%. Time above range (TAR) off camp was 53.4% and that on camp was 42.4%. The mean absolute relative difference (MARD) off camp was 19.7% ± 25.2%, whereas that on camp was 16.0% ± 14.8% (P = 0.367). Clinically acceptable zones of the error grid analyses were approximately 96% in total. Rt-CGM exhibited higher usability and recommendation scores in HCPs than those in campers. This may be related to relatively lower accuracy in rt-CGM. Overall usability and recommendation are clinically satisfactory, but due to relatively low accuracy, no decision should be made based on a single, non-verified SG value alone.

Research on Fault Prediction of Power Devices in Rod Control Power Cabinets Based on BiTCN-Attention Transfer Learning Model

The Insulated Gate Bipolar Transistor (IGBT) is the key power device in the rod control power cabinet of nuclear power plants; its reliable operation is of great significance for ensuring the safe and economical operation of the nuclear power plants. Therefore, it is necessary to conduct fault prediction research on IGBT to achieve better condition-based maintenance and improve its operational reliability. However, power cabinets often operate under multiple, complex working conditions, so predicting IGBT faults from single working condition data usually has limitations and low accuracy. Its failure probability has an important relationship with the actual operating conditions of the cabinet. In order to improve the reliability and maintainability of the control power cabinet in nuclear power plants, this paper takes IGBTs in the rod control power cabinet as the object and makes full use of the data of IGBTs under multiple working conditions to carry out research on the cross-condition fault prediction of IGBTs under multiple-source working conditions. A transfer learning (TL) model based on a bidirectional time convolutional network (BiTCN) combined with attention was proposed to solve the problem of low accuracy of cross-operating fault prediction in a multi-source domain. Firstly, an IGBT fault simulation model was built to collect the life cycle state data of the module under different working conditions. Then, after pre-processing such as removing outliers, kernel principal component analysis (KPCA) was used to integrate all source domain data, obtain source domain characterization data, and train the BiTCN-attention model. Finally, the BiTCN-attention model trained in the source domain was transferred, and the model was fine-tuned according to the target domain data. Simulation results show that the accuracy of the proposed BiTCN-attention transfer learning prediction method can reach more than 99%, which is significantly better than that of the recurrent neural network transfer learning (RNN-TL) model, long short-term memory network transfer learning (LSTM-TL) model, gated cyclic unit transfer learning (GRU-TL) model, and time convolutional network transfer learning (TCN-TL) model. This method can not only reduce the inconsistency of fault characteristic values caused by changes in working conditions but also accurately predict the degradation trend when only early fault data are available, providing an effective solution for IGBT fault prediction across working conditions in multi-source domains.

Utilizing Computer Vision and Deep Learning to Detect and Monitor Insects in Real Time by Analyzing Camera Trap Images

Insect monitoring techniques are often labor-intensive and need significant resources for identifying species after manual field traps. Insect traps are usually maintained every week, leading to a low temporal accuracy of information collected that impedes ecological analysis. This study introduces a handheld computer vision device to attract and detect real insects. The research explicitly proposes identifying and categorizing species by imaging live species drawn to a camera trapping. An Automatic Moth Trapping (AMT) equipped with light elemnets and a camera was developed to draw and observe insects throughout twilight and nocturnal periods. Moth Classification and Counting (MCC) utilizes Computer Vision (CV) and Deep Learning (DL) evaluation of collected pictures and monitors. It enumerates insect populations while identifying moth species. Over 48 nights, more than 250k photos were captured, averaging 5.6k daily. A tailored Convolutional Neural Networks (CNN) was developed on 2000 labeled photos of live insects across eight distinct categories. The suggested computer vision method and methodology have shown encouraging outcomes as an economical option for automated surveillance of insects.

Analyzing Soil Pollution by Image Processing and Machine Learning at Contaminated Agricultural Field

Due to the fast advancement of big data, applying Machine Learning (ML) techniques to detect Soil Pollution (SP) at Potentially Contaminated Sites (PCS) across many sectors and regional sizes has emerged as a prominent research focus. The challenges in acquiring essential indices of SP sources and routes result in present methodologies exhibiting low predictive accuracy and an inadequate scientific foundation. This study gathered environmental data concerning heavy metal and organic contamination from 200 PCS across six representative sectors. Twenty-one indices derived from fundamental data, potential SP from products and materials, SP efficacy, and the migrating capability of SP were employed to build the SP detection index method. The research integrated the score into the new characteristic group, including 11 indicators using consolidation computation. The newly selected feature subset was utilized for training ML designs, including Random Forests (RF), Support Vector Machines (SVM), and Multilayer Perceptrons (MLP), and evaluated to ascertain its impact on SP recognition methods. The study findings indicated that the four newly developed indices by feature fusion exhibit an association with SP comparable to that of the original index. The component analysis suggests that several indices related to fundamental information, contamination potential from products and raw materials, and SP prevention levels significantly influence SP to varying extents. The index of the migratory capability of soil contaminants has minimal influence on the classification job of SP detection inside PCS. This research introduces a novel technological approach for identifying SP via big data and ML techniques while offering an overview and scientific foundation for PCS's environmental administration and SP mitigation.

A Blockchain and Zero Knowledge Proof Based Data Security Transaction Method in Distributed Computing

In distributed computing, data trading mechanisms are essential for ensuring the sharing of data across multiple computing nodes. Nevertheless, they currently encounter considerable obstacles, including low accuracy in matching trading parties, ensuring fairness in transactions, and safeguarding data privacy throughout the trading process. In order to address these issues, we put forward a data trading security scheme based on zero-knowledge proofs and smart contracts. In the phase of preparing the security parameters, the objective is to reduce the complexity of generating non-interactive zero-knowledge proofs and to enhance the efficiency of data trading. In the pre-trading phase, we devise attribute atomic matching smart contracts based on precise data property alignment, with the objective of achieving fine-grained matching of data attributes between trading parties. In the trading execution phase, lightweight cryptographic algorithms based on elliptic curve cryptography (ECC) and non-interactive zero-knowledge proofs are employed for the dual encryption of trading data and the generation of attribute proof contracts, thus ensuring the security and privacy of the data. The results of experiments conducted on the Ethereum platform in an industrial IoT scenario demonstrate that our scheme maintains stable and low-cost consumption while ensuring accuracy in matching and privacy protection.

Decoding Brain Signals from Rapid-Event EEG for Visual Analysis Using Deep Learning

The perception and recognition of objects around us empower environmental interaction. Harnessing the brain’s signals to achieve this objective has consistently posed difficulties. Researchers are exploring whether the poor accuracy in this field is a result of the design of the temporal stimulation (block versus rapid event) or the inherent complexity of electroencephalogram (EEG) signals. Decoding perceptive signal responses in subjects has become increasingly complex due to high noise levels and the complex nature of brain activities. EEG signals have high temporal resolution and are non-stationary signals, i.e., their mean and variance vary overtime. This study aims to develop a deep learning model for the decoding of subjects’ responses to rapid-event visual stimuli and highlights the major factors that contribute to low accuracy in the EEG visual classification task.The proposed multi-class, multi-channel model integrates feature fusion to handle complex, non-stationary signals. This model is applied to the largest publicly available EEG dataset for visual classification consisting of 40 object classes, with 1000 images in each class. Contemporary state-of-the-art studies in this area investigating a large number of object classes have achieved a maximum accuracy of 17.6%. In contrast, our approach, which integrates Multi-Class, Multi-Channel Feature Fusion (MCCFF), achieves a classification accuracy of 33.17% for 40 classes. These results demonstrate the potential of EEG signals in advancing EEG visual classification and offering potential for future applications in visual machine models.

A microwave temperature sounder (MWTS)–microwave humidity sounder fusion cloud detection method for land data from the FengYun‐3D MWTS and its impact on error estimation

AbstractThe presence of clouds greatly disrupting the estimation of observation errors in satellite data. Therefore, cloud detection is a necessary step in satellite radiance data assimilation, not only to identify clear sky data and cloudy data, but also, more importantly, to accurately specify observation errors in different scenarios during all‐sky data assimilation. However, the assimilation of the microwave temperature sounder‐2 (MWTS‐2) onboard the FengYun‐3D satellite has been restricted by the low accuracy of cloud detection due to the absence of low‐frequency channels, especially for land areas where complex underlying surface brings more difficulties to the identification of cloudy data. The microwave humidity sounder‐2 (MWHS‐2), which is designed for water vapor sounding, is usually equipped on the same satellite as MWTS‐2. The sensitivity of the MWHS‐2 to water vapor enables it to provide more accurate information about clouds over land. Based on the advantage of the two instruments measuring on the same platform, an effective cloud detection method for MWTS‐2 data over land is established in this study after collocating the data of the MWTS‐2 and MWHS‐2. The results show that the new method achieves a cloud detection rate exceeding 75%, with a significantly reduced false detection rate compared with the traditional scattering index method, which is decreased from 40% to approximately 26%. More importantly, in non‐precipitating cloud regions, the cloud index built by the new method has a significant linear correlation with the difference between observed and simulated brightness temperatures (OMB) of MWTS‐2 data. Statistical analysis reveals that the cloud index can be used to effectively rectify the systematic OMB bias caused by those clouds in non‐precipitation cloud areas, which demonstrates the promising application of this cloud index in all‐sky MWTS‐2 data assimilation over land.