Search Results (13)

Task-oriented medical dialogue systems face two fundamental challenges: the explosion of state-action space caused by numerous diseases and symptoms and the sparsity of informative signals during interactive diagnosis. These issues significantly hinder the accuracy and efficiency of automated clinical reasoning. To address these problems, we propose MA-HRL, a multi-agent hierarchical reinforcement learning framework that decomposes the diagnostic task into specialized agents. A high-level controller coordinates symptom inquiry via multiple worker agents, each targeting a specific disease group, while a two-tier disease classifier refines diagnostic decisions through hierarchical probability reasoning. To combat sparse rewards, we design an information entropy-based reward function that encourages agents to acquire maximally informative symptoms. Additionally, medical knowledge graphs are integrated to guide decision-making and improve dialogue coherence. Experiments on the SymCat-derived SD dataset demonstrate that MA-HRL achieves substantial improvements over state-of-the-art baselines, including +7.2% diagnosis accuracy, +0.91% symptom hit rate, and +15.94% symptom recognition rate. Ablation studies further verify the effectiveness of each module. This work highlights the potential of hierarchical, knowledge-aware multi-agent systems for interpretable and scalable medical diagnosis. Full article

To address the issues of fragmented species information, low knowledge extraction efficiency, and insufficient utilization in the aquaculture domain, the main objective of this study is to construct the first knowledge graph for the Golden Pomfret aquaculture field and optimize the named entity recognition (NER) methods used in the construction process. The dataset contains challenges such as long text processing, strong local context dependencies, and entity sample imbalance, which result in low information extraction efficiency, recognition errors or omissions, and weak model generalization. This paper proposes a novel named entity recognition model, LA-CANER (Local Attention-Category Awareness NER), which combines local attention mechanisms with category awareness to improve both the accuracy and speed of NER. The constructed knowledge graph provides significant scientific knowledge support to Golden Pomfret aquaculture workers. First, by integrating and standardizing multi-source information, the knowledge graph offers comprehensive and accurate data, supporting decision-making for aquaculture management. The graph enables precise reasoning based on disease symptoms, environmental factors, and historical production data, helping workers identify potential risks early and take preventive actions. Furthermore, the knowledge graph can be integrated with large models like GPT-4 and DeepSeek-R1. By providing structured knowledge and rules, the graph enhances the reasoning and decision-making capabilities of these models. This promotes the application of smart aquaculture technologies and enables precision farming, ultimately increasing overall industry efficiency. Full article

Healthcare 4.0 addresses modernization and digital transformation challenges, such as home-based care and precision treatments, by leveraging advanced technologies to enhance accessibility and efficiency. Semantic technologies, particularly knowledge graphs (KGs), have proven instrumental in representing interconnected medical data and improving clinical decision-support systems. We previously introduced a semantic framework to assist medical experts during patient interactions. Operating iteratively, the framework prompts medical experts with relevant questions based on patient input, progressing toward accurate diagnoses in time-constrained settings. It comprises two components: (a) a KG representing symptoms, diseases, and their relationships, and (b) algorithms that generate questions and prioritize hypotheses—a ranked list of symptom–disease pairs. An earlier extension enriched the KG with a symptom ontology, incorporating hierarchical structures and inheritance relationships to improve accuracy and question-generation capabilities. This paper further extends the framework by introducing strategies tailored to specific medical domains. Strategies integrate domain-specific knowledge and algorithms, refining decision making while maintaining the iterative nature of expert–patient interactions. We demonstrate this approach using an emergency medicine case study, focusing on life-threatening conditions. The KG is enriched with attributes tailored to emergency contexts and supported by dedicated algorithms. Boolean rules attached to graph edges evaluate to TRUE or FALSE at runtime based on patient-specific data. These enhancements optimize decision making by embedding domain-specific goal-oriented knowledge and inference processes, providing a scalable and adaptable solution for diverse medical contexts. Full article

Model robustness is an important index in medical cybersecurity, and hard-negative samples in electronic medical records can provide more gradient information, which can effectively improve the robustness of a model. However, hard negatives pose difficulties in terms of their definition and acquisition. To solve these problems, a data augmentation approach consisting of fusing SimBERT and a knowledge graph for application to a hard-negative sample is proposed in this paper. Firstly, we selected 40 misdiagnosed cases of diabetic complications as the original data for data augmentation. Secondly, we divided the contents of the electronic medical records into two parts. One part consisted of the core disease phrases in the misdiagnosed case records, which a medical specialist selected. These denoted the critical diseases that the model diagnosed as negative samples. Based on these core symptom words, new symptom phrases were directly generated using the SimBERT model. On the other hand, the noncore phrases of misdiagnosed medical records were highly similar to the positive samples. We determined the cosine similarity between the embedding vector of the knowledge graph entities and a vector made up of the noncore phrases. Then, we used Top-K sampling to generate text. Finally, combining the generated text from the two parts and the disturbed numerical indexes resulted in 160 enhancement samples. Our experiment shows that the distances between the samples generated using the SimKG-BERT model’s samples were closer to those of the positive samples and the anchor points in the space vector were closer than those created using the other models. This finding is more in line with how hard negatives are defined. In addition, compared with the model without data augmentation, the F1 values in the three data sets of diabetic complications increased by 6.4%, 2.24%, and 5.54%, respectively. The SimKG-BERT model achieves data augmentation in the absence of misdiagnosed medical records, providing more gradient information to the model, which not only improves the robustness of the model but also meets the realistic needs of assisted-diagnosis safety. Full article

Deep Reinforcement Learning is a key technology for the diagnosis-oriented medical dialogue system, determining the type of disease according to the patient’s utterances. The existing dialogue models for disease diagnosis cannot achieve good performance due to the large number of symptoms and diseases. In this paper, we propose a knowledge-enhanced hierarchical reinforcement learning model for strategy learning in the medical dialogue system for disease diagnosis. Our hierarchical strategy alleviates the problem of a large action space in reinforcement learning. In addition, the knowledge enhancement module integrates a learnable disease–symptom relationship matrix and medical knowledge graph into the hierarchical strategy for higher diagnosis success rate. Our proposed model has been proved to be effective on a medical dialogue dataset for automatic disease diagnosis. Full article

Most of the existing medical knowledge maps are incomplete and need to be completed/predicted to obtain a complete knowledge map. To solve this problem, we propose a knowledge graph embedding model (Cyclic_CKGE) based on cyclic consistency. The model first uses the “graph” constructed with the head entity and relationship to predict the tail entity, and then uses the “inverse graph” constructed with the tail entity and relationship to predict the head entity. Finally, the semantic space distance between the head entity and the original head entity should be very close, which solves the reversibility problem of the network. The Cyclic_CKGE model with a parameter of 0.46 M has the best results on FB15k-237, reaching 0.425 Hits@10. Compared with the best model R-GCN, its parameter exceeds 8 M and reaches 0.417 Hits@10. Overall, Cyclic_CKGE’s parametric efficiency is more than 17 times that of R-GCNs and more than 8 times that of DistMult. In order to better show the practical application of the model, we construct a visual medical information platform based on a medical knowledge map. The platform has three kinds of disease information retrieval methods: conditional query, path query and multi-symptom disease inference. This provides a theoretical method and a practical example for realizing knowledge graph visualization. Full article

Parkinson’s disease (PD) is a movement disorder caused by a dopamine deficit in the brain. Current therapies primarily focus on dopamine modulators or replacements, such as levodopa. Although dopamine replacement can help alleviate PD symptoms, therapies targeting the underlying neurodegenerative process are limited. The study objective was to use artificial intelligence to rank the most promising repurposed drug candidates for PD. Natural language processing (NLP) techniques were used to extract text relationships from 33+ million biomedical journal articles from PubMed and map relationships between genes, proteins, drugs, diseases, etc., into a knowledge graph. Cross-domain text mining, hub network analysis, and unsupervised learning rank aggregation were performed in SemNet 2.0 to predict the most relevant drug candidates to levodopa and PD using relevance-based HeteSim scores. The top predicted adjuvant PD therapies included ebastine, an antihistamine for perennial allergic rhinitis; levocetirizine, another antihistamine; vancomycin, a powerful antibiotic; captopril, an angiotensin-converting enzyme (ACE) inhibitor; and neramexane, an N-methyl-D-aspartate (NMDA) receptor agonist. Cross-domain text mining predicted that antihistamines exhibit the capacity to synergistically alleviate Parkinsonian symptoms when used with dopamine modulators like levodopa or levodopa–carbidopa. The relationship patterns among the identified adjuvant candidates suggest that the likely therapeutic mechanism(s) of action of antihistamines for combatting the multi-factorial PD pathology include counteracting oxidative stress, amending the balance of neurotransmitters, and decreasing the proliferation of inflammatory mediators. Finally, cross-domain text mining interestingly predicted a strong relationship between PD and liver disease. Full article

Problems of preeclampsia sign diagnosis are mostly based on symptom data with the characteristics of data collected periodically in uncertain, ambiguous, and obstetrician opinions. To reduce the effects of preeclampsia, many studies have investigated the disease, prevention, and complication. Conventional fuzzy inference techniques can solve several diagnosis problems in health such as fuzzy inference systems (FIS), and Mamdani complex fuzzy inference systems with rule reduction (M-CFIS-R), however, the computation time is quite high. Recently, the research direction of approximate inference based on fuzzy knowledge graph (FKG) has been proposed in the M-CFIS-FKG model with the combination of regimens in traditional medicine and subclinical data gathered from medical records. The paper has presented a proposed model of FKG-Pairs3 to support patients’ disease diagnosis, together with doctors’ preferences in decision-making. The proposed model has been implemented in real-world applications for disease diagnosis in traditional medicine based on input data sets with vague information, quantified by doctor’s preferences. To validate the proposed model, it has been tested in a real-world case study of preeclampsia signs in a hospital for disease diagnosis with the traditional medicine approach. Experimental results show that the proposed model has demonstrated the model’s effectiveness in the decision-making of preeclampsia signs. Full article

Semantic data integration provides the ability to interrelate and analyze information from multiple heterogeneous resources. With the growing complexity of medical ontologies and the big data generated from different resources, there is a need for integrating medical ontologies and finding relationships between distinct concepts from different ontologies where these concepts have logical medical relationships. Standardized Medical Ontologies are explicit specifications of shared conceptualization, which provide predefined medical vocabulary that serves as a stable conceptual interface to medical data sources. Intelligent Healthcare systems such as disease prediction systems require a reliable knowledge base that is based on Standardized medical ontologies. Knowledge graphs have emerged as a powerful dynamic representation of a knowledge base. In this paper, a framework is proposed for automatic knowledge graph generation integrating two medical standardized ontologies- Human Disease Ontology (DO), and Symptom Ontology (SYMP) using a medical online website and encyclopedia. The framework and methodologies adopted for automatically generating this knowledge graph fully integrated the two standardized ontologies. The graph is dynamic, scalable, easily reproducible, reliable, and practically efficient. A subgraph for cancer terms is also extracted and studied for modeling and representing cancer diseases, their symptoms, prevention, and risk factors. Full article

(1) Hypophosphatasia (HPP) is a rare inherited disease caused by mutations (pathogenic variants) in the ALPL gene which encodes tissue-nonspecific alkaline phosphatase (TNSALP). HPP is characterized by impaired bone mineral metabolism due to the low enzymatic activity of TNSALP. Knowledge about the structure of the gene and the features and functions of various ALPL gene variants, taking into account population specificity, gives an understanding of the hereditary nature of the disease, and contributes to the diagnosis, prevention, and treatment of the disease. The purpose of the study was to describe the spectrum and analyze the functional features of the ALPL gene variants, considering various HPP subtypes and clinical symptoms in Russian children. (2) From 2014–2021, the study included the blood samples obtained from 1612 patients with reduced alkaline phosphatase activity. The patients underwent an examination with an assessment of their clinical symptoms and biochemical levels of TNSALP. DNA was isolated from dried blood spots (DBSs) or blood from the patients to search for mutations in the exons of the ALPL gene using Sanger sequencing. The PCR products were sequenced using a reagent BigDye Terminator 3.1 kit (Applied Biosystems). Statistical analysis was performed using the GraphPad Prism 8.01 software. (3) The most common clinical symptoms in Russian patients with HPP and two of its variants (n = 22) were bone disorders (75%), hypomyotonia (50%), and respiratory failure (50%). The heterozygous carriage of the causal variants of the ALPL gene was detected in 225 patients. A total of 2 variants were found in 27 patients. In this group (n = 27), we identified 28 unique variants of the ALPL gene, of which 75.0% were missense, 17.9% were frameshift, 3.6% were splicing variants, and 3.6% were duplications. A total of 39.3% (11/28) of the variants were pathogenic, with two variants being probably pathogenic, and 15 variants had unknown clinical significance (VUS). Among the VUS group, 28.6% of the variants (7/28) were discovered by us for the first time. The most common variants were c.571G > A (p.Glu191Lys) and c.1171del (Arg391Valfs*12), with frequencies of 48.2% (13/28) and 11% (3/28), respectively. It was found that the frequency of nonsense variants of the ALPL gene was higher (p < 0.0001) in patients with the perinatal form compared to the infantile and childhood forms of HPP. Additionally, the number of homozygotes in patients with the perinatal form exceeded (p < 0.01) the frequencies of these genotypes in children with infantile and childhood forms of HPP. On the contrary, the frequencies of the compound-heterozygous and heterozygous genotypes were higher (p < 0.01) in patients with infantile childhood HPP than in perinatal HPP. In the perinatal form, residual TNSALP activity was lower (p < 0.0005) in comparison to the infantile and childhood (p < 0.05) forms of HPP. At the same time, patients with the heterozygous and compound-heterozygous genotypes (mainly missense variants) of the ALPL gene had greater residual activity (of the TNSALP protein) regarding those homozygous patients who were carriers of the nonsense variants (deletions and duplications) of the ALPL gene. Residual TNSALP activity was lower (p < 0.0001) in patients with pathogenic variants encoding the amino acids from the active site and the calcium and crown domains in comparison with the nonspecific region of the protein. Full article

The research aim is to construct a disease-symptom knowledge graph (DSKG) as a cause-effect knowledge graph containing disease-symptom relations as a cause-effect relation type determined from downloaded documents on medical web-board resources. Each disease-symptom relation connects a disease-name concept node (a causative-concept node) to a corresponding node having a group of correlated symptom-concept/effect-concept features as common symptom-concept/effect-concept features among some disease-name concepts. The DSKG benefits non-professionals in preliminary diagnosis through a recommender web-board. There are three main problems: how to determine symptom concepts from sentences without annotation on the documents having disease-name concepts as the documents’ topic-names; how to determine the disease-symptom relations from the documents with/without complications; and how to construct the DSKG involving high dimensional symptom-concept features after union of the correlated symptom-concept groups. Therefore, we apply a word co-occurrence pattern including medical-symptom expressions from Wikipedia including MeSH and the Lexitron Dictionary to determine the symptom concepts. The Cartesian product is applied for automatic-supervised machine learning to determine the disease-symptom relation. We propose using Principal Component Analysis for constructing the DSKG by dimensionality reduction in the symptom-concept features with minimized information loss. In contrast to previous works, the proposed approach enables the DSKG construction with precise and concise representation scores of 7.8 and 9, respectively. Full article

COVID-19 and heart failure (HF) are common disorders and although they share some similar symptoms, they require different treatments. Accurate diagnosis of these disorders is crucial for disease management, including patient isolation to curb infection spread of COVID-19. In this work, we aim to develop a computer-aided diagnostic system that can accurately differentiate these three classes (normal, COVID-19 and HF) using cough sounds. A novel handcrafted model was used to classify COVID-19 vs. healthy (Case 1), HF vs. healthy (Case 2) and COVID-19 vs. HF vs. healthy (Case 3) automatically using deoxyribonucleic acid (DNA) patterns. The model was developed using the cough sounds collected from 241 COVID-19 patients, 244 HF patients, and 247 healthy subjects using a hand phone. To the best our knowledge, this is the first work to automatically classify healthy subjects, HF and COVID-19 patients using cough sounds signals. Our proposed model comprises a graph-based local feature generator (DNA pattern), an iterative maximum relevance minimum redundancy (ImRMR) iterative feature selector, with classification using the k-nearest neighbor classifier. Our proposed model attained an accuracy of 100.0%, 99.38%, and 99.49% for Case 1, Case 2, and Case 3, respectively. The developed system is completely automated and economical, and can be utilized to accurately detect COVID-19 versus HF using cough sounds. Full article

There has long been an interest in understanding how the hazards from spaceflight may trigger or exacerbate human diseases. With the goal of advancing our knowledge on physiological changes during space travel, NASA GeneLab provides an open-source repository of multi-omics data from real and simulated spaceflight studies. Alone, this data enables identification of biological changes during spaceflight, but cannot infer how that may impact an astronaut at the phenotypic level. To bridge this gap, Scalable Precision Medicine Oriented Knowledge Engine (SPOKE), a heterogeneous knowledge graph connecting biological and clinical data from over 30 databases, was used in combination with GeneLab transcriptomic data from six studies. This integration identified critical symptoms and physiological changes incurred during spaceflight. Full article