Natural Language Processing Research Articles

Application of Transformers to Chemical Synthesis

Efficient chemical synthesis is critical for the production of organic chemicals, particularly in the pharmaceutical industry. Leveraging machine learning to predict chemical synthesis and improve the development efficiency has become a significant research focus in modern chemistry. Among various machine learning models, the Transformer, a leading model in natural language processing, has revolutionized numerous fields due to its powerful feature-extraction and representation-learning capabilities. Recent applications demonstrated that Transformer models can also significantly enhance the performance in chemical synthesis tasks, particularly in reaction prediction and retrosynthetic planning. This article provides a comprehensive review of the applications and innovations of Transformer models in the qualitative prediction tasks of chemical synthesis, with a focus on technical approaches, performance advantages, and the challenges associated with applying the Transformer architecture to chemical reactions. Furthermore, we discuss the future directions for improving the applications of Transformer models in chemical synthesis.

Lenspeak: Text to Speech Glasses

This research explores the creation of "Text-to- Speech (TTS) Glasses," an innovative assistive device designed to enhance accessibility for individuals with visual impairments. The glasses facilitate the real-time transformation of printed text into audible speech, empowering users to independently access textual information. Utilizing a Raspberry Pi as he processing core, the system integrates advanced Optical Character Recognition (OCR) and Text-to-Speech (TTS) technologies. This paper delves into the system’s architecture, functionality, benefits, limitations, and prospective advancements eywords—Text-to-Speech (TTS) Raspberry Pi Speech Syn- thesis Embedded Systems Speech Processing Voice Synthesis Natural Language Processing (NLP) Open-source TTS Audio Output

PALLM: Evaluating and Enhancing PALL iative Care Conversations with L arge L anguage M odels

Effective patient-provider communication is crucial in clinical care, directly impacting patient outcomes and quality of life. Traditional evaluation methods, such as human ratings, patient feedback, and provider self-assessments, are often limited by high costs and scalability issues. Although existing natural language processing (NLP) techniques show promise, they struggle with the nuances of clinical communication and require sensitive clinical data for training, reducing their effectiveness in real-world applications. Emerging large language models (LLMs) offer a new approach to assessing complex communication metrics, with the potential to advance the field through integration into passive sensing and just-in-time intervention systems. This study explores LLMs as evaluators of palliative care communication quality, leveraging their linguistic, in-context learning, and reasoning capabilities. Specifically, using simulated scripts crafted and labeled by healthcare professionals, we test proprietary models (e.g., GPT-4) and fine-tune open-source LLMs (e.g., LLaMA2) with a synthetic dataset generated by GPT-4 to evaluate clinical conversations, to identify key metrics such as ‘understanding’ and ‘empathy’. Our findings demonstrated LLMs’ superior performance in evaluating clinical communication, providing actionable feedback with reasoning, and demonstrating the feasibility and practical viability of developing in-house LLMs. This research highlights LLMs’ potential to enhance patient-provider interactions and lays the groundwork for downstream steps in developing LLM-empowered clinical health systems.

VIEWER: an extensible visual analytics framework for enhancing mental healthcare.

A proof-of-concept study aimed at designing and implementing Visual & Interactive Engagement With Electronic Records (VIEWER), a versatile toolkit for visual analytics of clinical data, and systematically evaluating its effectiveness across various clinical applications while gathering feedback for iterative improvements. VIEWER is an open-source and extensible toolkit that employs natural language processing and interactive visualization techniques to facilitate the rapid design, development, and deployment of clinical information retrieval, analysis, and visualization at the point of care. Through an iterative and collaborative participatory design approach, VIEWER was designed and implemented in one of the United Kingdom's largest National Health Services mental health Trusts, where its clinical utility and effectiveness were assessed using both quantitative and qualitative methods. VIEWER provides interactive, problem-focused, and comprehensive views of longitudinal patient data (n = 409870) from a combination of structured clinical data and unstructured clinical notes. Despite a relatively short adoption period and users' initial unfamiliarity, VIEWER significantly improved performance and task completion speed compared to the standard clinical information system. More than 1000 users and partners in the hospital tested and used VIEWER, reporting high satisfaction and expressed strong interest in incorporating VIEWER into their daily practice. VIEWER provides a cost-effective enhancement to the functionalities of standard clinical information systems, with evaluation offering valuable feedback for future improvements. VIEWER was developed to improve data accessibility and representation across various aspects of healthcare delivery, including population health management and patient monitoring. The deployment of VIEWER highlights the benefits of collaborative refinement in optimizing health informatics solutions for enhanced patient care.

AWS Lambda and SageMaker: Real-Time Solutions for Machine Learning

Natural Language Processing (NLP) has emerged as a transformative power during the era of significant data. The vast and diverse vocabulary generated by big data allows for the development of advanced NLP models through machine learning algorithms and distributed computing methods. The collaboration of NLP and big data has led to the creation of strong language models such as BERT and GPT, enhancing NLP's capability to comprehend content and provide insights across a range of applications, including reasoning, machine translation, automated responses, and personalized NLP text generation. Integrating NLP within a big data framework addresses numerous challenges across various industries. Utilizing data collection, business intelligence can provide real-time insights, while collaboration can be enhanced through improved networking. Sentiment analysis aids in refining product and market research by enabling organizations to understand customer opinions and preferences. Cloud services like AWS Lambda simplify serverless execution of NLP workflows, facilitating cost-effective, scalable, and event-driven processing of large datasets. Moreover, AWS SageMaker streamlines the development, training, and deployment of machine learning models, enabling the efficient creation and operationalization of advanced NLP solutions on a large scale. This study demonstrates the efficacy of NLP in processing large datasets, particularly for sentiment analysis using the MapReduce framework, alongside the operational advantages offered by AWS Lambda and SageMaker. In conclusion, this document emphasizes the opportunities and challenges that arise from the integration of NLP with big data and cloud-native tools, providing insights into their combined capabilities. This potential can be harnessed in various ways, fostering a deeper understanding of both language and data analysis.

Resume QR Generator (The helper AI)

This research paper is about to job seekers The AI Resume Analyzer is a tool that uses Natural Language Processing (NLP) to make the hiring process faster and easier. In today's job market, companies receive a large number of resumes, making it difficult and timeconsuming to find the best candidates. This AI-based system solves this problem by automatically reviewing and ranking resumes, saving time and effort for employers.The tool works by extracting text from resumes, no matter the format—PDF, Word, or plain text. Using NLP, it identifies and organizes important details such as contact information, skills, work experience, education, and other key sections. It also checks how well a candidate's resume matches the job requirements. Factors like relevant skills, quality of experience, and use of industry-specific keywords are taken into account. By ranking resumes based on these factors, the AI Resume Analyzer helps companies quickly find qualified candidates. It makes recruitment more efficient, reduces bias, and can handle a large number of applications, making it a great solution for businesses of all sizes. Key Words: Hiring process, job application ,recruitment, Natural language processing

P1241 Identification and characterization of patients with Crohn's disease and complex perianal fistulas using Natural Language Processing and machine learning: INTUITION-CPF Study.

Abstract Background Crohn’s disease (CD) is an inflammatory condition that can cause perianal fistulas (mostly complex), affecting 26-35% of CD patients. There is no gold-standard treatment for perianal disease, leading to significant unmet needs. This study uses Artificial Intelligence (AI) to analyze electronic medical records (EMRs) to identify and characterize CD patients with complex perianal fistulas (CPF). It offers an overview of patient characteristics, treatment patterns, and CPF incidence in CD patients attending inflammatory bowel disease units in Spain. Methods INTUITION-CPF is a national, multicenter, retrospective study. EMRs of patients with CD, diagnosed or not with CPF (overall CD population) from January 2015-December 2021, were reviewed and processed using an AI analysis based on machine learning and Natural Language Processing (NLP). There were two phases in the study. Phase 1: NLP-identified patients with CD+CPF were manually validated to assess sensitivity and specificity; and Phase 2: characteristics, treatments and visits to departments were analysed in eligible patients with CD+/-CPF. Here, we present results from the phase 2. Results A total of 2,182 patients with CD were identified. Median age (range) was 51 years (21-105). 263 patients (12%) suffered from CPF, of which 70% experienced pain, 48% active (suppurative) disease, and 28% perianal abscesses. The estimated annual incidence of CPF among CD patients was 1.6%. Patients with CD+CPF had more medical treatments (91%) compared to overall CD patients (74%). Proportion of patients receiving biologics was near double for CD+CPF compared to overall CD patients (41% vs 21%) with infliximab the most frequently used (26%) (Table 1). 13% of the overall CD population had ≥1 surgery, compared with 36% of patients with CD+CPF. The main surgeries in patients with CD+CPF included seton placement (14%), advancement flap surgery (FLAP) (16%) and ostomy (13%) (Table 1). Hospitalisation (≥24h) was required for 49% of overall CD patients vs 64% of those with CPF. Gastroenterology (73%) was the most frequently visited department, followed by surgery (31%) and emergency departments (27%) (Figure 1). Conclusion This study demonstrates that CD patients with CPF require more complex and multidisciplinary management, including more biological treatments and surgeries, and exhibit higher hospitalization and emergency visit rates compared to overall CD patients.Inconsistencies with the existing literature highlight the opportunity to further enhance NLP techniques to achieve more accurate outcomes.

P0788 Exploring topics of Dutch Inflammatory Bowel Disease patient messages for improving transmural communication efficiency: A natural language processing approach

Abstract Background Patients with inflammatory bowel disease (IBD) in the Netherlands included in a remote monitoring care path can use the myIBDcoach app to communicate with their healthcare providers1. Currently, IBD physicians and nurses respond to the messages without triage, leading to limited communication efficiency and increased workload. We aim to identify common topics in Dutch IBD patient messages with natural language processing (NLP) topic models to enable automated triage. Methods We analyzed 13,199 text messages from 856 Dutch IBD patients sent via myIBDcoach between September 23, 2014, and October 19, 2022. After anonymization and cleaning, the messages were split into 42,537 sentences and processed using two Dutch text-embedding models: multilingual Sentence-Transformer and BERTje.2,3 Model-generated topics were manually merged and mapped to 16 predefined labels (medical or administrative) developed with expert input (Table 1). Messages were categorized as medical if they contained at least one sentence-level medical topic; otherwise, they were categorized as administrative. To evaluate model performance, we also manually annotated a random sample of 328 sentences from 100 messages. Model performance was assessed using label-matching accuracy, defined as the percentage of matches between computer-generated labels and manual annotations. Descriptive statistics were used to analyze topic distribution and message categories. Results The model with multilingual Sentence-Transformer embeddings (average label-matching accuracy = 50%) outperformed BERTje (average label-matching accuracy = 43%) and was chosen. Excluding irrelevant information (n = 8,722) such as greetings, the five most frequently identified sentence-level topics were Test Procedure & Results (n = 5,968), Appointment/Contact (n = 5,784), Symptoms (n = 4,690), Medication (n = 4,361), and General Health Update (n = 3,472). The two most common topics were administrative. The distribution of all identified sentence-level topics can be found in Figure 1. After aggregating the sentence-level topics back to the original 13,199 messages, we found that a significant proportion (41%, n = 5,412) contained exclusively administrative topics. In contrast, 59% of the messages (n = 7,787) concerned at least one medical topic requiring specialist responses. Conclusion Using NLP, we identified the common topics in text messages from Dutch IBD patients and evaluated the performance of general Dutch topic models in this specialized context. The findings of our research provide a strong basis for developing automated triage systems to increase response efficiency and reduce provider workload in the future for digital monitoring of IBD.

Automated Detection of Cancer-Suspicious Findings in Japanese Radiology Reports with Natural Language Processing: A Multicenter Study.

Missed critical imaging findings, particularly those indicating cancer, are a common issue that can result in delays in patient follow-up and treatment. To address this, we developed a rule-based natural language processing (NLP) algorithm to detect cancer-suspicious findings from Japanese radiology reports. The dataset used consisted of chest and abdomen CT reports from six institutions. Reports from our institution were used for algorithm development and internal evaluation, while reports from the other five institutions were used for external evaluation. To create the gold standard, reports were annotated by two experienced physicians. Data were statistically analyzed using precision, recall and F1 score with 1000 bootstrap iterations. BERT was used as a baseline deep learning model, and its performance was compared with the proposed rule-based method. At the report level of detection, the overall precision, recall, and F-1 score were 0.886, 0.886, and 0.883, respectively, for the rule-based algorithm, which were higher than those of the deep learning algorithm (0.851, 0.679, and 0.733). The overall results include both internal and external validation data. For the internal validation set, the precision, recall, and F-1 score were 0.929, 0.929, and 0.927, respectively. For the external validation set, the precision, recall, and F-1 score were 0.875, 0.879, and 0.873, demonstrating generalizability. In conclusion, we show the rule-based NLP algorithm exhibited a high performance in detecting cancer-suspicious findings from multi-institutional CT reports.

Deep Multi-View Clustering Optimized by Long Short-Term Memory Network

Long short-term memory (LSTM) networks have shown great promise in sequential data analysis, especially in time-series and natural language processing. However, their potential for multi-view clustering has been largely underexplored. In this paper, we introduce a novel approach called deep multi-view clustering optimized by long short-term memory network (DMVC-LSTM), which leverages the sequential modeling capability of LSTM to effectively integrate multi-view data. By capturing complex interdependencies and nonlinear relationships between views, DMVC-LSTM improves clustering accuracy and robustness. The method includes three feature fusion techniques—concatenation, averaging, and attention-based fusion—with concatenation as the primary method. Notably, DMVC-LSTM is well suited for datasets that exhibit symmetry, as it can effectively handle symmetrical relationships between views while preserving the underlying structures. Extensive experiments demonstrate that DMVC-LSTM outperforms existing multi-view clustering algorithms, particularly in high-dimensional and complex datasets, achieving superior performance in datasets like 20 Newsgroups and Wikipedia Articles. This paper presents the first application of LSTM in multi-view clustering, marking a significant step forward in both clustering performance and the application of LSTM in multi-view data analysis.

Dendrites endow artificial neural networks with accurate, robust and parameter-efficient learning

Artificial neural networks (ANNs) are at the core of most Deep Learning (DL) algorithms that successfully tackle complex problems like image recognition, autonomous driving, and natural language processing. However, unlike biological brains who tackle similar problems in a very efficient manner, DL algorithms require a large number of trainable parameters, making them energy-intensive and prone to overfitting. Here, we show that a new ANN architecture that incorporates the structured connectivity and restricted sampling properties of biological dendrites counteracts these limitations. We find that dendritic ANNs are more robust to overfitting and match or outperform traditional ANNs on several image classification tasks while using significantly fewer trainable parameters. These advantages are likely the result of a different learning strategy, whereby most of the nodes in dendritic ANNs respond to multiple classes, unlike classical ANNs that strive for class-specificity. Our findings suggest that the incorporation of dendritic properties can make learning in ANNs more precise, resilient, and parameter-efficient and shed new light on how biological features can impact the learning strategies of ANNs.

Longitudinal evolution of the transdiagnostic prodrome to severe mental disorders: a dynamic temporal network analysis informed by natural language processing and electronic health records.

Modelling the prodrome to severe mental disorders (SMD), including unipolar mood disorders (UMD), bipolar mood disorders (BMD) and psychotic disorders (PSY), should consider both the evolution and interactions of symptoms and substance use (prodromal features) over time. Temporal network analysis can detect causal dependence between and within prodromal features by representing prodromal features as nodes, with their connections (edges) indicating the likelihood of one feature preceding the other. In SMD, node centrality could reveal insights into important prodromal features and potential intervention targets. Community analysis can identify commonly occurring feature groups to define SMD at-risk states. This retrospective (2-year) cohort study aimed to develop a global transdiagnostic SMD network of the temporal relationships between prodromal features and to examine within-group differences with sub-networks specific to UMD, BMD and PSY. Electronic health records (EHRs) from South London and Maudsley(SLaM) NHS Foundation Trust were included from 6462 individuals with SMD diagnoses (UMD:2066; BMD:740; PSY:3656). Validated natural language processing algorithms extracted the occurrence of 61 prodromal features every three months from two years to six months before SMD onset. Temporal networks of prodromal features were constructed using generalised vector autoregression panel analysis, adjusting for covariates. Edge weights (partial directed correlation coefficients, z) were reported in autocorrelative, unidirectional and bidirectional relationships. Centrality was calculated as the sum of (non-autoregressive) connections leaving (out-centrality, cout) or entering (in-centrality, cin) a node. The three sub-networks (UMD, BMD, PSY) were compared using permutation analysis, and community analysis was performed using Spinglass. The SMD network revealed strong autocorrelations (0.04 ≤ z ≤ 0.10), predominantly positive connections, and identified aggression (cout = 0.103) and tearfulness (cin = 0.134) as the most central features. Sub-networks for UMD, BMD, and PSY showed minimal differences, with 3.5% of edges differing between UMD and PSY, 0.8% between UMD and BMD, and 0.4% between BMD and PSY. Community analysis identified one positive psychotic community (delusional thinking-hallucinations-paranoia) and two behavioural communities (aggression-cannabis use-cocaine use-hostility, aggression-agitation-hostility) as the most common. This study represents the most extensive temporal network analysis conducted on the longitudinal interplay of SMD prodromal features. The findings provide further evidence to support transdiagnostic early detection services across SMD, refine assessments to detect individuals at risk and identify central features as potential intervention targets.

Augmenting LLMs to Securely Retrieve Information for Construction and Facility Management

In the past few years, generative AI has seen remarkable progress. The emergence of the transformer architecture has facilitated the creation of highly advanced language models that generate text, summarize content, and translate languages with impressive accuracy. Our study introduces a retrieval-augmented generation system tailored to the dynamic needs of facility management. The proposed system aims to provide instant, accurate access to essential information by integrating advanced techniques from natural language processing and information retrieval paradigms. The implementation leverages the Mixtral 8x7B model for multilingual text processing and the Milvus vector database for efficient document storage and retrieval. The dataset used includes documents such as images, operation manuals, inspection results, blueprints, and technical drawings, in various file formats. This diverse dataset reflects the variety of information encountered in construction and facility management. The evaluation involved generating question–answer pairs pertinent to facility management tasks and assessing the system’s performance using metrics such as ROUGE, BLEU, and semantic similarity. The findings suggest that retrieval-augmented generation systems can significantly enhance operational efficiency by reducing the time and effort required to access information while maintaining high security and data privacy standards.

P0554 Machine Learning-Based Interpretation of Unstructured Radiology Reports for Objective Scoring of Inflammation and Stenosis in IBD: A Nationwide Study From the epi-IIRN

Abstract Background Systematic interpretation and quantification of inflammation and stenosis in inflammatory Bowel Disease (IBD) from radiology reports of CT and MR enterography exams are challenging due to their unstructured nature, often leading to variability and reduced reproducibility. This study leverages data from the epi-IIRN nationwide cohort to develop and evaluate machine-learning-based scores for inflammation and stenosis derived from these unstructured radiology reports. Methods We identified radiology reports for IBD-diagnosed patients within the nationwide epi-IIRN cohort, which encompasses data from Israel’s four HMOs, covering 98% of the population. Using our in-house HSMP-BERT natural language processing (NLP) platform, we extracted radiological indicators related to inflammation, stenosis, and location, including wall thickening, enhancement, lumen narrowing, and dilation across specified regions: jejunum, ileum, caecum, colon, sigmoid, and rectum. We developed and evaluated a machine-learning model to predict both categorical (no, mild, and moderate or severe) and VAS (0-100) scores of global inflammation and stenosis, using physician-based scoring as the reference standard. A 5-fold cross-validation experimental setup was applied to assess model performance. Evaluation metrics included accuracy, F1 score, Cohen’s kappa, AUC, PPV, and NPV for categorical scores, and RMSE, r2 and MAE for VAS-like scores. Results The dataset included 9658 reports from 7389 patients with a mean age of 36.4±17.6 years, of whom 49% were male. In the annotated subset (500 reports), inflammation severity was distributed as approximately 37% no, 34% mild, and 29% moderate or severe, while stenosis severity was distributed as 72% no, 15% mild, and 13% moderate or severe. Table 1 summarizes key metrics for inflammation and stenosis, including accuracy, PPV, NPV, F1, and Cohen’s kappa for categorical scores, and RMSE, r2, and MAPE for VAS-like scores, averaged over 5 folds with mean [95% CI]. Main results showed strong performance for both inflammation and stenosis predictions: for inflammation, mean accuracy/F1/kappa/AUC/NPV/PPV/r2/RMSE/MAE were 0.786/0.782/0.676/0.914/0.897/0.775/0.629/13.57/8.571, and for stenosis these values reached 0.912/0.906/0.793/0.955/0.959/0.837/0.760/8.431/3.270. Figure 1 illustrates the joint distribution of predicted versus ground truth VAS-like scores, with both axes spanning the full range of 0 to 100 as defined by physician-based scoring. Conclusion This study demonstrates the feasibility of machine-learning models for mass scoring of inflammation and stenosis from unstructured radiology reports of IBD patients, enabling reliable mass tagging of CTE and MRE as well as determining disease phenotype in large populations.

Automated extraction of post-stroke functional outcomes from unstructured electronic health records.

Population level tracking of post-stroke functional outcomes is critical to guide interventions that reduce the burden of stroke-related disability. However, functional outcomes are often missing or documented in unstructured notes. We developed a natural language processing (NLP) model that reads electronic health records (EHR) notes to automatically determine the modified Rankin Scale (mRS). We included consecutive patients (⩾18 years) with acute stroke admitted to our center (2015-2024). mRS scores were obtained from the Get With the Guidelines registry and clinical notes (if documented), and used as the gold standard to compare against NLP-generated scores. We used text-based features from notes, along with age, sex, discharge status, and outpatient follow-up to train a logistic regression for prediction of good (0-2) versus poor (3-6) mRS, and a linear regression for the full range of mRS scores. The models were trained for prediction of mRS at hospital discharge and post-discharge. The models were externally validated in a dataset of patients with brain injuries from a different healthcare center. We included 5307 patients, 5006 in train and test and 301 in validation; average age was 69 (SD 15) and 65 (SD 17) years, respectively; 47% female. The logistic regression achieved an area under the receiver operating curve (AUROC) of 0.94 [CI 0.93-0.95] (test) and 0.94 [0.91-0.96] (validation), and the linear model a root mean squared error (RMSE) of 0.91 [0.87-0.94] (test) and 1.17 [1.06-1.28] (validation). The NLP-based model is suitable for use in large-scale phenotyping of stroke functional outcomes and population health research.

Построение графа знаний по телекоммуникационным данным

The paper presents a method for building a knowledge graph from telecommunication data using proprietary and ref-erence models commonly employed in the telecommunications domain. Reference models are based on those included in the framework developed by the TM Forum consortium. The proposed approach involves building a knowledge graph for proprietary models through automated processing of autotest log files and billing system database tables. The relevance of knowledge graphs stems from their structured and semantic nature, as well as their potential for applying machine learning algorithms to generate recommendations for optimizing telecommunication processes and systems. A method based on a multi-step reasoning approach is proposed for creating interpretable recommendations by predict-ing and restoring missing links in a proprietary knowledge graph. This approach treats multi-step reasoning as a ques-tion-answering task using natural language processing techniques. The implementation of the proposed solution, lever-aging a transformer-based neural network architecture, yielded interpretable results while maintaining metric values comparable to existing methods.

Deciphering abbreviations in Malaysian clinical notes using machine learning.

This is the first Malaysian machine learning model to detect and disambiguate abbreviations in clinical notes. The model has been designed to be incorporated into MyHarmony, a Natural Language Processing system, that extracts clinical information for healthcare management. The model utilizes word embedding to ensure feasibility of use, not in real-time but for secondary analysis, within the constraints of low-resource settings. A Malaysian clinical embedding, based on Word2Vec model, was developed using 29,895 electronic discharge summaries. The embedding was compared against conventional rule-based and FastText embedding on two tasks: abbreviation detection and abbreviation disambiguation. Machine Learning classifiers were applied to assess performance. The Malaysian clinical word embedding contained 7 million word-tokens, 24,352 unique vocabularies, and 100 dimensions. For abbreviation detection, the Decision Tree classifier augmented with the Malaysian clinical embedding showed the best performance (F-score of 0.9519). For abbreviation disambiguation, the classifier with the Malaysian clinical embedding had the best performance for most of the abbreviations (F-score of 0.9903). Despite having a smaller vocabulary and dimension, our local clinical word embedding performed better than the larger non-clinical FastText embedding. Word embedding with simple machine learning algorithms can decipher abbreviations well. It also requires lower computational resources and is suitable for implementation in low-resource settings such as Malaysia. The integration of this model into MyHarmony will improve recognition of clinical terms, thus improving the information generated for monitoring Malaysian healthcare services and policymaking.

Database Query and Reasoning

The rapid evolution of data-driven decision-making has highlighted the need for systems capable of seamlessly bridging the gap between natural human language and structured query languages like SQL. Traditional database systems often demand a technical understanding of SQL syntax and database schemas, creating significant barriers for non-technical users. To address this challenge, the "Database Query and Reasoning" project introduces an intelligent system designed to interpret natural language inputs, convert them into accurate SQL queries, and deliver both query results and explanatory reasoning for the outputs. This approach empowers users, regardless of technical expertise, to interact with complex databases effectively and derive actionable insights. Keywords: Natural Language Processing (NLP),AI Model Integration, CSV-to-Database Conversion, Reasoning Module,Query Result Visualization

Artificial Intelligence (AI) and Liquid Biopsy Transforming Early Detection of Liver Metastases in Gastrointestinal Cancers.

Liver metastases from Gastrointestinal (GI) cancers present significant challenges in oncology, often signaling poor prognosis. Traditional detection methods like imaging and tissue biopsies have limitations in sensitivity, specificity, and tumor heterogeneity represen-tation. The advent of artificial intelligence (AI) in healthcare, driven by advancements in ma-chine learning, algorithms, and data science, offers a promising frontier for early detection and management of liver metastases. This review explores the integration of AI and liquid biopsy technologies as transformative tools in the proactive detection of liver metastases aris-ing from GI malignancies. Liquid biopsy, a non-invasive method, analyzes circulating tumor cells (CTCs), cell-free DNA (cfDNA), and circulating tumor DNA (ctDNA) in bodily fluids. It provides a compre-hensive overview of tumor heterogeneity and enables real-time monitoring of tumor evolu-tion and treatment response. Despite its advantages, liquid biopsy faces challenges such as low sensitivity for early-stage metastases, reduced detectability due to liver filtration, and technical limitations. AI enhances the potential of liquid biopsies by improving diagnostic accuracy through ad-vanced algorithms like Convolutional Neural Networks (CNNs) and Natural Language Pro-cessing (NLP). These AI models analyze complex biomedical data, offering higher sensitivity and specificity in cancer detection. The synergy between AI and liquid biopsies promises early detection, better disease monitoring, and personalized treatment strategies. This review underscores the significant advancements AI and liquid biopsy technologies bring to oncological precision medicine, particularly in improving overall survival (OS) and disease-free survival (DFS) for patients with GI cancer metastases. As we transition into the era of precision medicine, the integration of these technologies holds the potential to redefine cancer care and patient management.

Development and Implementation of Automated Referral Triaging System for Spinal Cord Stimulation Procedure in Pain Medicine.

Effective referral triaging enhances patient service outcomes, experience and access to care especially for specialized procedures. This study presents the development and implementation of an automated triaging system to predict patients who would benefit from Spinal Cord Stimulation (SCS) procedure for their pain management. The proposed triage system aims to improve the triage process by reducing unnecessary appointments before SCS assessment, ensuring appropriate pain management care. It compares various machine learning techniques for the prediction while addressing the class imbalance and overlap challenges inherent in the data. Both data-level and algorithm-level approaches were explored. Two years of patient data was collected including patient characteristics, diagnosis history, pain symptoms, appointment history, medication history, and concepts from clinical notes extracted using Natural Language Processing. EasyEnsemble with Ada Boosting method, an algorithm-level approach, showed the most promising results. The tenfold validation indicated the average area under curve of 0.82, true positive rate (TPR) of 77.3%, and true negative rate (TNR) of 73.0%. The probability threshold was adjusted to 0.575 to meet practice expectation of 15% or less on false positive rate (FPR). The implementation pipeline for the selected model was designed to be applicable to real clinical settings. The one-year implementation results showed TPR of 64.7% and TNR of 87.2%, which reduced FPR by 12.8% while reduced TPR by 12.6%. The trade-off was acceptable to practice. The proposed triage system demonstrated promising accuracy, leading to the enhancement of scheduling systems, patient care, and the reduction of unnecessary appointments in a pain medicine setting.