Personalized Medicine Data Analysis

Heterogeneous datasets are pervasive today, existing in various domains. Objects within these complex datasets are often represented from different perspectives, at different scales, or through multiple modalities, such as images, sensor readings, language sequences, and compact mathematical statements. Such datasets have been analyzed in the past using Multi-View Learning (MVL), Multi-Task Learning (MTL), and Tensor Learning (TL). In recent years, Multi-Modal Learning (MML) has also been employed. MML is a Machine Learning (ML) approach that integrates and processes information from multiple types of data, with different "perspectives" or "modalities" such as text, images, audio, video, or sensor data. The goal of MML is to leverage the complementary strengths of these modalities to improve model performance and enable richer understanding and predictions. Precision medicine and personalized clinical decision support systems (CDSS) tools have long aimed to leverage multimodal patient data to better capture complex, high-dimensional patient states and provider responses. This data ranges from free-form text notes and semi-structured electronic health records (EHR) to high-frequency physiological signals. While the advent of transformer architectures has enabled deeper insights from merging modalities, it has also required meticulous feature engineering and alignment. In patient monitoring, effectively analyzing diverse physiological signals within CDSS is highly challenging. #MedicalInformatics To address the challenges of analyzing multimodal patient data, the authors of [1] introduce MedTsLLM, a general multimodal large language model (LLM) framework that effectively integrates time series data and rich contextual information in the form of text. This framework performs three clinically relevant tasks (in time-series) which enable deeper analysis of physiological signals and can provide actionable insights for clinicians: • semantic segmentation • boundary detection • anomaly detection At a high level, boundary detection splits signals into periods like breaths or beats. Semantic segmentation further splits time series into distinct, meaningful segments. Anomaly detection identifies periods within the signals that deviate from normal. MedTsLLM utilizes a reprogramming layer to align embeddings of time series patches with a pretrained LLM's embedding space, making effective use of raw time series in conjunction with textual context. They additionally tailored the text prompt to include patient-specific information. Their experiments showed that MedTsLLM outperforms state-of-the-art baselines, including deep learning models, other LLMs, and clinical methods, across multiple medical domains, specifically electrocardiograms (ECG) and respiratory waveforms. Links to their preprint [1] and #Python GitHub repository [2] are shared in the comments.

🟥 Integrative Omics for Drug Response Prediction One of the biggest challenges facing modern medicine is predicting how individual patients will respond to specific therapies. While traditional clinical metrics can provide some guidance, they often fail to capture the molecular complexity of each patient's disease. Integraomics—a powerful approach that combines genomic, transcriptomic, proteomic, epigenomic, and metabolomic data—is rapidly emerging as a disruptive force in the field of personalized drug response prediction. By analyzing multiple molecular levels simultaneously, integraomics can provide a systems-level understanding of how a patient's biology interacts with therapeutic drugs. For example, genomic variants may reveal potential drug targets, while transcriptomic profiles can show whether those targets are actively expressed. Proteomic and metabolomic data can provide further insights into functional consequences and metabolic vulnerabilities. Together, these datasets help clinicians match patients with the most effective drugs and avoid treatments that are likely to fail or produce adverse reactions. Recent advances include artificial intelligence and machine learning-driven platforms that leverage integraomics data to predict individual drug responses with high accuracy. These models can identify gene expression patterns, signaling pathway activity, and metabolic signatures that are associated with sensitivity or resistance to targeted therapies, chemotherapy, or immunotherapy. In addition, integrative omics has also shown great value in real-time monitoring, allowing clinicians to adjust treatment plans based on the progression of a patient's tumor or changes in immune response over time. This dynamic approach can not only improve efficacy, but also reduce unnecessary toxicity and medical costs. In summary, integrative omics is ushering in a new era of precision medicine - drug selection is driven by deep molecular insights and tailored to each patient's unique biological characteristics. It is transforming treatment from reactive to predictive treatment, and from one-size-fits-all to truly personalized medicine. References [1] Hui-O Chen et al., J Pers Med 2024 (https://lnkd.in/eADqTPDS) [2] Ruijiang Li et al., Advanced Intelligent Systems 2024 (https://lnkd.in/eDNafrtz) #IntegrativeOmics #DrugResponsePrediction #PrecisionMedicine #MultiOmics #AIinHealthcare #PersonalizedTherapy #TranslationalResearch #OmicsData #Genomics #Proteomics #Metabolomics #CancerTherapy #MachineLearning #SystemsBiology #FutureOfMedicine #CSTEAMBiotech

Identifying cancer-related mutations accurately is a critical step in precision medicine. Today, we’ve published new research in Nature Biotechnology on 🧬DeepSomatic🧬, an AI-powered tool that uses machine learning to identify genetic variants, or mutations, in cancer cells more accurately than current methods. This work is aimed at helping researchers pinpoint what's driving a cancer and informing more effective treatment plans. Somatic variant detection is an integral part of cancer genomics analysis. While most methods have focused on short-read sequencing, long-read technologies offer potential advantages to discover variants in the hardest to sequence parts of the genome. 🧬 About the model:  DeepSomatic was rigorously trained on high-confidence data, a feat made possible by working with our partners at UC Santa Cruz. The model is capable of accurately differentiating actual genetic cancer variants from the technical artifacts introduced during sample preservation, addressing a critical hurdle in early detection. 🧬 Superior Accuracy and Clinical Impact:  DeepSomatic consistently outperformed other tools across all major sequencing platforms. It shows major improvements in identifying complex insertions and deletions (Indels). Furthermore, in a new study with partners at Children's Mercy, DeepSomatic successfully found ten small variants in pediatric leukemia cells that were missed by other tools. 🧬 Flexible and Broad Use:  The model is flexible, working across all major sequencing platforms, and can be applied to both tumor-normal and challenging tumor-only samples, extending its utility for complex cancer types. 🧬 Open Access:  We are making DeepSomatic and the CASTLE dataset openly available to the research community. DeepSomatic is the most recent addition to our 10-year journey developing open source methods for geneticists to study the genomes of humans, plants, and animals. We are excited to see how researchers and drug manufacturers will use these resources to develop more effective, personalized treatments for cancer patients. The ability to accurately identify these subtle genetic drivers is key to unlocking new therapies. More in our blog authored by Kishwar Shafin and Andrew Carroll: https://goo.gle/4n23gIB   Read the full article in Nature Biotechnology: https://lnkd.in/drxii8fz

Is RWE + RCT As Close to A ‘Perfect Marraige’ As One Can Get?? The Indispensable Value of Hybrid Data in Advancing Psychiatry Combining Randomized Controlled Trial (RCT) data with Real-World Evidence (RWE) is not just beneficial—it's absolutely crucial for truly advancing psychiatric treatment and achieving personalized medicine. RCTs have long been the gold standard, providing high internal validity to establish the efficacy of an intervention. They isolate variables, control for confounders, and demonstrate causality under meticulously managed conditions. This is essential for drug approval and initial clinical guidelines. However, they operate in a highly selective environment. The patient populations in RCTs are often homogeneous, excluding individuals with common comorbidities, polypharmacy use, or greater severity, which are the very characteristics of patients seen daily in clinical practice. This is where RWE steps in. Drawn from diverse sources like electronic health records (EHRs), patient registries, insurance claims data, and even passive monitoring via wearables, RWE provides a robust measure of effectiveness and safety in the wild. It reflects the true patient journey: complex medication adherence patterns, varied clinician interpretations, and the impact of social determinants of health. Bridging the Efficacy-Effectiveness Gap In psychiatry, where conditions are inherently heterogeneous (e.g., depression, bipolar disorder) and treatment responses are highly variable, this hybrid approach is transformative. Integrating RCT data (RWE) with RWE (RCT) allows us to: * Understand Treatment Response Variability: We can use RWE to identify clinical and genetic subgroups that respond optimally to an intervention initially proven efficacious in an RCT, moving us closer to truly personalized care. * Assess Long-Term Safety and Tolerability: While RCTs typically run for a fixed duration, RWE offers invaluable, longitudinal data on adverse event profiles and persistence of treatment effects over months or years, which is critical for chronic mental health conditions. * Validate and Generalize Findings: RWE validates RCT findings in broader, more representative populations, ensuring that a treatment deemed "effective" is actually helping the majority of patients outside of a research setting. Let's champion this data synergy to move beyond one-size-fits-all care and build smarter, more patient-centered mental health solutions. This collaborative approach between researchers, clinicians, and data scientists will ultimately translate to better outcomes for patients facing complex psychiatric disorders.

🔍 Could Pharmacogenomics (PGx) Transform Medication Management in Long-Term Care? In my previous role as VP of Pharmacy at Precision Health Solutions, I had the opportunity to implement PGx testing across long-term care (LTC) facilities in Florida, where the impact was both profound and eye-opening. Here’s a snapshot of 6 facilities that we worked with and what we uncovered: - 91 residents were identified by providers as needing PGx testing due to polypharmacy, specific disease states, and medication needs. - After testing, I categorized each resident into one of four risk levels—Immediate Attention, High Risk, Moderate Risk, and Low Risk—to support providers in translating PGx data into actionable steps. - The results were significant: 55% of the residents (50 individuals) fell into the Immediate Attention and High Risk categories, indicating a strong need for personalized medication plans. -Lastly, there were 116 Drug-Gene Interactions identified among those 50 patients, and 211 deprescribing or medication adjustment opportunities. PGx testing provided us with critical insights that would have otherwise gone unnoticed, underscoring the need for precision medicine in LTC. For consultant pharmacists and providers, these tools can lead to better, safer patient outcomes and more effective care. Let’s continue the conversation on how PGx can drive safer, more personalized medication strategies in long-term care settings. #Pharmacogenomics #PrecisionMedicine #LongTermCare #Polypharmacy #ConsultantPharmacist #SeniorCare #PatientSafety