arXiv subjects

Meysam Ghaffari

Publications and source records attributed to Meysam Ghaffari.

12 recordsLinked to original sources

arXiv2026

Structured Evidence Routing for Incident Risk Prediction from Multimodal Longitudinal EHRs

Incident risk prediction from longitudinal electronic health records (EHRs) is challenging because relevant signals are multimodal, weak in isolation, and distributed across irregular patient histories. We propose structured evidence routing, a router-predictor-reviewer workflow that separates full-record access from disease-specific assessment. The router organizes the complete pre-index EHR into a compact summary and targeted evidence slices; the predictor uses this evidence to form an evidence-linked risk assessment, which the reviewer critiques. For comparison with supervised EHRSHOT baselines, we pair the routed evidence summaries with a supervised classifier readout. Across five 1-year incident diagnosis tasks, our method reaches the AUROC range of established supervised EHRSHOT baselines and remains competitive on AUPRC, while exposing a patient-specific evidence trail. Internal pre-readout ablations further suggest that routing, laboratory evidence, task guidance, and review each contribute to performance.

cs.AI↗

arXiv2026

Foundation Agents Meet Agentic Deep Research: Evidence-Grounded Clinical Code Forecasting

Next-encounter ICD forecasting predicts which standardized diagnosis codes will be documented at a future visit from the longitudinal record available beforehand. The task is prospective and multi-label: the target note does not yet exist, and several codes may be correct. Structured EHR foundation models capture recurrence and temporal progression, whereas language foundation models generate flexible diagnostic hypotheses. We introduce ICD-Deepresearch, a DeepResearch workflow that composes these predictive foundation models with medical search and ICD dictionaries. Because no source reveals the future code set, research evaluates candidate transitions by linking patient evidence, external clinical relations, and exact code semantics under a fixed top-K budget. Candidate Generation uses SparseEHR to produce an EHR Prior that initializes two bounded Research Expansion rounds; an independent GPT-5 Direct Forecast supplies complementary candidates. Final Selection validates, deduplicates, and jointly ranks both paths, after which a separate module writes rationales without changing predictions. Finally ICD-Deepresearch achieves patient-averaged precision/recall of 24.60/35.09% on MIMIC-III and 25.14/48.32% on MIMIC-IV. Physicians rate 51% and 68% of its retrieved documents useful, compared with 22% and 39% for standalone GPT-5 web search and 32% and 41% for Medical Deep Research. ICD-Deepresearch therefore improves over the registered local comparators while retrieving evidence with higher physician-rated usefulness than the standalone research systems

cs.CL↗

arXiv2026

Reasoning Quality Emerges Early: Data Curation for Reasoning Models

Supervised fine-tuning (SFT) on a small, high-quality set of long reasoning traces is an effective approach for eliciting strong reasoning capabilities in Large Language Models (LLMs). However, existing methods for curating high-quality SFT data rely heavily on strong reasoning models to filter examples based on diversity and difficulty, making the curation process costly while often yielding suboptimal data quality. In this work, we show that diverse and challenging reasoning examples can be identified using only the initial reasoning tokens. Specifically, we demonstrate that difficult problems can be reliably detected based on the loss of the first 100 reasoning tokens evaluated at a randomly perturbed checkpoint of the pretrained model. We further show that examples exhibiting similar loss patterns over their first 1k reasoning tokens across a small number of perturbed checkpoints extrapolating along the fine-tuning trajectory provably induce similar gradients. We validate our approach through extensive experiments on fine-tuning Qwen2.5-7B and Llama3.1-8B models on the M23K medical reasoning and OpenThoughts-Math datasets. Our method outperforms existing baselines by up to 1.7% while being 91% more token efficient.

cs.LG↗

arXiv2026

Text Over Image: Auditing Multimodal Robustness in Synthetic Medical Image Detection

With the rapid adoption of generative AI, synthetic medical images pose growing risks, including diagnostic deception and insurance fraud. Although prior work has explored vision-language model (VLM)-based synthetic image detection, these evaluations typically consider images in isolation. In clinical practice, however, images are interpreted alongside structured records and metadata, and VLMs are increasingly deployed under joint image-record inputs. We uncover a previously underexamined multimodal vulnerability: when given both modalities, VLMs may overweight record context in authenticity judgments, such that the same image receives different predictions solely due to changes in its accompanying text. This raises concerns about robustness in real-world deployment. To systematically characterize this effect, we reformulate synthetic medical image detection as an audit of multimodal robustness at the image-record interface and introduce a paired benchmark that holds the image fixed while swapping controlled metadata variants. Across multiple imaging modalities, we evaluate diverse open-weight and frontier API VLMs and find that changing the metadata context alone can flip authenticity judgments, with accuracy on authentic images dropping by 61.1% on average under an explicit AI-origin tag. We further propose an inference-time mitigation pipeline that detects and neutralizes provenance shortcuts without model retraining, substantially outperforming direct prompt-based suppression on the affected subset. Our benchmark provides a standardized tool for assessing and improving multimodal robustness beyond image-only settings. Code and data will be released upon acceptance.

cs.CV↗

arXiv2019

Next-Generation High-Resolution Vector-Borne Disease Risk Assessment

Vector-borne diseases cause more than 1 million deaths annually. Estimates of epidemic risk at high spatial resolutions can enable effective public health interventions. Our goal is to identify the risk of importation of such diseases into vulnerable cities at the granularity of neighborhoods. Conventional models cannot achieve such spatial resolution, especially in real-time. Besides, they lack real-time data on demographic heterogeneity, which is vital for accurate risk estimation. Social media, such as Twitter, promise data from which demographic and spatial information could be inferred in real-time. On the other hand, such data can be noisy and inaccurate. Our novel approach leverages Twitter data, using machine learning techniques at multiple spatial scales to overcome its limitations, to deliver results at the desired resolution. We validate our method against the Zika outbreak in Florida in 2016. Our main contribution lies in proposing a novel approach that uses machine learning on social media data to identify the risk of vector-borne disease importation at a sufficiently fine spatial resolution to permit effective intervention. This will lead to a new generation of epidemic risk assessment models, promising to transform public health by identifying specific locations for targeted intervention.

cs.SI↗

arXiv2019

High-resolution home location prediction from tweets using deep learning with dynamic structure

Timely and high-resolution estimates of the home locations of a sufficiently large subset of the population are critical for effective disaster response and public health intervention, but this is still an open problem. Conventional data sources, such as census and surveys, have a substantial time lag and cannot capture seasonal trends. Recently, social media data has been exploited to address this problem by leveraging its large user-base and real-time nature. However, inherent sparsity and noise, along with large estimation uncertainty in home locations, have limited their effectiveness. Consequently, much of previous research has aimed only at a coarse spatial resolution, with accuracy being limited for high-resolution methods. In this paper, we develop a deep-learning solution that uses a two-phase dynamic structure to deal with sparse and noisy social media data. In the first phase, high recall is achieved using a random forest, producing more balanced home location candidates. Then two deep neural networks are used to detect home locations with high accuracy. We obtained over 90% accuracy for large subsets on a commonly used dataset. Compared to other high-resolution methods, our approach yields up to 60% error reduction by reducing high-resolution home prediction error from over 21% to less than 8%. Systematic comparisons show that our method gives the highest accuracy both for the entire sample and for subsets. Evaluation on a real-world public health problem further validates the effectiveness of our approach.

cs.SI↗

arXiv2017

Duplicate matching and estimating features for detection of copy-move images forgery

Copy-move forgery is the most popular and simplest image manipulation method. In this type of forgery, an area from the image copied, then after post processing such as rotation and scaling, placed on the destination. The goal of Copy-move forgery is to hide or duplicate one or more objects in the image. Key-point based Copy-move forgery detection methods have five main steps: preprocessing, feature extraction, matching, transform estimation and post processing that matching and transform estimation have important effect on the detection. More over the error could happens in some steps due to the noise. The existing methods process these steps separately and in case of having an error in a step, this error could be propagated to the following steps and affects the detection. To solve the above mentioned problem, in this paper the steps of the detection system interact with each other and if an error happens in a step, following steps are trying to detect and solve it. We formulate this interaction by defining and optimizing a cost function. This function includes matching and transform estimation steps. Then in an iterative procedure the steps are executed and in case of detecting error, the error will be corrected. The efficiency of the proposed method analyzed in diverse cases such as pixel image precision level on the simple forgery images, robustness to the rotation and scaling, detecting professional forgery images and the precision of the transformation matrix. The results indicate the better efficiency of the proposed method.

cs.MM↗

arXiv2016

P4QS: A Peer to Peer Privacy Preserving Query Service for Location-Based Mobile Applications

The location-based services provide an interesting combination of cyber and physical worlds. However, they can also threaten the users' privacy. Existing privacy preserving protocols require trusted nodes, with serious security and computational bottlenecks. In this paper, we propose a novel distributed anonymizing protocol based on peer-to-peer architecture. Each mobile node is responsible for anonymizing a specific zone. The mobile nodes collaborate in anonymizing their queries, without the need not get access to any information about each other. In the proposed protocol, each request will be sent with a randomly chosen ticket. The encrypted response produced by the server is sent to a particular mobile node (called broker node) over the network, based on the hash value of this ticket. The user will query the broker to get the response. All parts of the messages are encrypted except the fields required for the anonymizer and the broker. This will secure the packet exchange over the P2P network. The proposed protocol was implemented and tested successfully, and the experimental results showed that it could be deployed efficiently to achieve user privacy in location-based services.

cs.CR↗

arXiv2016

BigFCM: Fast, Precise and Scalable FCM on Hadoop

Clustering plays an important role in mining big data both as a modeling technique and a preprocessing step in many data mining process implementations. Fuzzy clustering provides more flexibility than non-fuzzy methods by allowing each data record to belong to more than one cluster to some degree. However, a serious challenge in fuzzy clustering is the lack of scalability. Massive datasets in emerging fields such as geosciences, biology and networking do require parallel and distributed computations with high performance to solve real-world problems. Although some clustering methods are already improved to execute on big data platforms, but their execution time is highly increased for large datasets. In this paper, a scalable Fuzzy C-Means (FCM) clustering named BigFCM is proposed and designed for the Hadoop distributed data platform. Based on the map-reduce programming model, it exploits several mechanisms including an efficient caching design to achieve several orders of magnitude reduction in execution time. Extensive evaluation over multi-gigabyte datasets shows that BigFCM is scalable while it preserves the quality of clustering.

cs.DC↗

arXiv2015

Improving image watermarking based on Tabu search by Chaos

With the fast development of communication and multimedia technology, the rights of the owners of multimedia products is vulnerable to the unauthorized copies and watermarking is one of the best known methods for proving the ownership of a product. In this paper we prosper the previous watermarking method which was based on Tabu search by Chaos. The modification applied in the permutation step of watermarking and the initial population generation of the Tabu search. We analyze our method on some well known images and experimental results shows the improvement in the quality and speed of the proposed watermarking method.

cs.MM↗

arXiv2014

Generating Graphical Chain by Mutual Matching of Bayesian Network and Extracted Rules of Bayesian Network Using Genetic Algorithm

With the technology development, the need of analyze and extraction of useful information is increasing. Bayesian networks contain knowledge from data and experts that could be used for decision making processes But they are not easily understandable thus the rule extraction methods have been used but they have high computation costs. To overcome this problem we extract rules from Bayesian network using genetic algorithm. Then we generate the graphical chain by mutually matching the extracted rules and Bayesian network. This graphical chain could shows the sequence of events that lead to the target which could help the decision making process. The experimental results on small networks show that the proposed method has comparable results with brute force method which has a significantly higher computation cost.

cs.AI↗

arXiv2014

Ambiguity-Driven Fuzzy C-Means Clustering: How to Detect Uncertain Clustered Records

As a well-known clustering algorithm, Fuzzy C-Means (FCM) allows each input sample to belong to more than one cluster, providing more flexibility than non-fuzzy clustering methods. However, the accuracy of FCM is subject to false detections caused by noisy records, weak feature selection and low certainty of the algorithm in some cases. The false detections are very important in some decision-making application domains like network security and medical diagnosis, where weak decisions based on such false detections may lead to catastrophic outcomes. They are mainly emerged from making decisions about a subset of records that do not provide enough evidence to make a good decision. In this paper, we propose a method for detecting such ambiguous records in FCM by introducing a certainty factor to decrease invalid detections. This approach enables us to send the detected ambiguous records to another discrimination method for a deeper investigation, thus increasing the accuracy by lowering the error rate. Most of the records are still processed quickly and with low error rate which prevents performance loss compared to similar hybrid methods. Experimental results of applying the proposed method on several datasets from different domains show a significant decrease in error rate as well as improved sensitivity of the algorithm.

cs.AI↗