• 学术搜索
  • 科研智能体
    • Research Labs
    • AI 阅读
    • AI 文库
    • 深度研究
    • 学者亮点
  • 学术资源
    • AI2000
    • 期刊/会议
    • 学者库
    • 学术API
      MCP&Skill
    • 溯源树
    • 数据集
  • 知识沉淀
    • 学术空间
订阅小程序
旧版功能
aminer vip
开通会员低至0.73元/天
一次搞定AI科研
立即登录
  • English
  • 联系方式
    O

    Optum

    企业
    1,558论文总数
    1.6万引用总数

    is an American pharmacy benefit manager and health care provider. It is a subsidiary of UnitedHealth Group since 2011. UHG formed Optum by merging its existing pharmacy and care delivery services into the single Optum brand, comprising three main businesses: OptumHealth, OptumInsight and OptumRx. In 2017, Optum accounted for 44 percent of UnitedHealth Group's profits and as of 2019, Optum's revenues have surpassed $100 billion. Also in early 2019, Optum gained significant media attention regarding a trade secrets lawsuit that the company filed against former executive David William Smith, after Smith left Optum to join Haven, the joint healthcare venture of Amazon, JPMorgan Chase, and Berkshire Hathaway. On January 15, 2019, Optum said that the revenue of the company had surpassed $100 billion for the first time, growing by 11.1% year over year to $101.3 billion. This also means that Optum will contribute more than 50% of UnitedHealth's 2020 earnings, making Optum UnitedHealth's fastest-growing unit.

    论文量&引用量时间轴

    机构学者

    论文(1558)

    年份
    起
    –
    止
    排序
    1Clinical and Cost Outcomes in Medicare Advantage Following Post-Acute Care
    Elan Gada, Paul Pangburn, Chris Sahr, Megan Jarvis, Richard Young

    Introduction: Post-acute care (PAC) disposition decisions influence costs and outcomes. Skilled nursing facilities (SNFs) provide intensive oversight, home health agencies (HHA) deliver care in familiar settings, often at lower cost. Comparative effectiveness for clinically similar Medicare Advantage members remains unclear. The purpose of this article is to compare clinical and cost outcomes for Medicare Advantage members discharged to HHA versus SNF after acute hospitalizations using a retrospective, quasi-experimental framework. Methods: Using deidentified medical claims from care organizations across 7 states, we identified members discharged to HHA or SNF following acute hospitalizations. Propensity score matching balanced demographics, clinical characteristics, prior utilization, and social determinants of health. Outcomes included readmissions, mortality, costs, emergency room visits, and primary care utilization. Results: No significant difference in 30-day readmissions (5.9% vs 6.4%, OR 0.91, 95% CI 0.67-1.25) or 90-day mortality (8.2% vs 9.9%, OR 0.81, 95% CI 0.64-1.02). HHA had lower costs at 30 days (-$7510, P < .001). Conclusions: Among Medicare Advantage members with comparable clinical profiles, discharge to home with HHA yields similar clinical outcomes at substantially lower cost than SNF. These savings present opportunities to reinvest in transitional care programs and address social determinants of health within the home setting. Findings support consideration of home-based care as a cost-effective alternative for appropriate members.

    2026HOME HEALTH CARE MANAGEMENT AND PRACTICE(2026)引用:21
    引用
    AI阅读
    加入学术空间
    2Utilizing Machine Learning for Predicting PrEP Use Status among Sexual and Gender Minority Young Adults
    Callisto Boka, Elizabeth A. Yonko, Mehrab Beikzadeh, Kimmo Kärkkäinen, Chenglin Hong, Majid Sarrafzadeh,Ian W. Holloway

    Pre-exposure prophylaxis (PrEP) is a highly effective biomedical prevention tool for HIV yet remains underutilized among key populations, particularly among young sexual and gender minorities (SGM). Recognizing the popularity of specific dating and social media apps among SGM young adults, we leveraged user data from these platforms to build a machine learning (ML) model that could inform targeted, data-driven interventions aimed at improving PrEP uptake and adherence. We adapted eWellness, an Android mobile app, to passively collect data from research participants capturing mobile app usage, keystroke patterns and logs, and GPS location data between 2021 and 2024. These data were used to train a ML model to predict self-reported PrEP use. Model accuracy was evaluated through F1 scores across different data types and feature combinations. Study protocols were developed in collaboration with community partners and adhered to strict ethical and privacy standards. A total of 82 SGM young adults participated, with 46 (56%) reporting PrEP use at baseline. Our machine learning model demonstrated good predictive accuracy for predicting PrEP use and non-use, achieving an F1 score of 0.84 (PrEP use) and 0.82 (non-use) outcomes when incorporating data from all mobile apps, including messaging, dating, and social media mobile apps. By contrast, predictions based solely on social media mobile app usage, language associated with sexual behavior and substance use risk, or location monitoring demonstrated worse accuracy (F1 scores of 0.79/0.75, 0.70/0.57, and 0.70/0.52, respectively). Additional feature extraction methods, as well as various combinations of these features, were also tested. However, none achieved predictive accuracy as well as the model incorporating all mobile app usage data combined. This study demonstrates the potential of machine learning to accurately predict PrEP use status among SGM young adults. The findings offer a foundation for developing more personalized PrEP promotion strategies, particularly among SGM young adults who use social media and dating apps. Future research should assess the model's adaptability across diverse SGM subgroups to further inform intervention development. Registry: ClinicalTrials.gov, ID: NCT04710901, November 9, 2020.

    2026Prevention Science(2026)引用:2
    引用
    AI阅读
    加入学术空间
    3AI-Augmented Block-Sketch Hybrid Compression: Enabling Semantic Query Processing on Compressed Data
    Srinubabu Kilaru, Venkata Subba Rao Yarlagadda,Madhusudan Bangalore Nagaraja

    The exponential growth of enterprise, scientific, and archival data demands systems that are both storage-efficient and query-efficient. Traditional compression algorithms reduce storage footprint but severely limit query performance, as full decompression is typically required before search operations. This paper introduces AI-Augmented Block-Sketch Hybrid Compression (AI-BSHC), a novel framework that integrates artificial intelligence into the compression-query pipeline. The proposed system partitions data into blocks and compresses them using standard algorithms, while each block is annotated with lightweight sketches for fast filtering and AI-driven embeddings for semantic retrieval. At query time, sketches rapidly eliminate irrelevant blocks, embeddings provide semantic similarity scoring, and only top-ranked candidates undergo decompression. We present the complete architecture, algorithms, and mathematical analysis of AI-BSHC, discuss its implementation using Zstandard compression, Bloom filters, and BERT embeddings, and evaluate its performance on diverse datasets including log files, scientific sensor data, and digital research archives. Experimental results demonstrate compression ratios of 0.33-0.38, query latency reductions of $3-6 \times$, and semantic recall exceeding 85 %. This work demonstrates how AI can fundamentally enhance compressed storage systems, enabling intelligent and context-aware querying at scale.

    20262026 IEEE 16th Annual Computing and Communication Workshop and Conference (CCWC)(2026)引用:1
    引用
    AI阅读
    加入学术空间
    4Context Poisoning As Extreme-Value Attention Interference in Long-Context Language Models
    Meysam Ghaffari, Nina Fatehi, Bhaskar Sen, Nasim Sabetpour, Carlos Morato

    Large language models can process increasingly long prompts, yet their ability to locate and use decisive evidence may degrade as irrelevant or confusable context is added. We formulate this phenomenon, which we call context poisoning, as extreme-value interference in attention: the decisive-evidence score is upper-bounded, while the maximum score among effective distractors grows with their number. Under a softmax retrieval abstraction, we derive a finite-sample upper bound showing that maintaining a fixed accuracy target above base rate requires the evidence margin to scale as Ω(√(log N)), where N denotes the effective distractor count rather than necessarily the raw context length. The analysis connects long-context degradation to score aliasing, positional aliasing, and softmax dilution. Controlled experiments show that retrieval accuracy decreases as total context grows in the presence of embedded hard negatives, that the same-format condition produces the largest observed accuracy drop among the tested distractor constructions at fixed context length, and that retrieval gating can improve evidence use while its net benefit depends on preserving evidence recall. These results motivate evidence bottlenecks, alias-resistant representations, retrieve-then-reason architectures, verifier-mediated memory, and contrastive anti-poison training.

    2026引用:1
    引用
    AI阅读
    加入学术空间
    5Clinical Outcomes Following Hematopoietic Stem Cell Transplant-Associated Thrombotic Microangiopathy in the United States: the Endure Study
    Michelle Long Schoettler, Andrew Messali, Kirti Batra, Arpita Nag, Yogesh Patel, Qiana Amos,Christine S. Higham

    Objectives Thrombotic microangiopathy (TMA) is a serious and potentially life-threatening complication of hematopoietic stem cell transplantation (HSCT), and as previously presented, associated with increased healthcare resource utilization. This analysis examined differences in cumulative incidence of early and late clinical complications between HSCT recipients with and without TMA. Methods The ENDURE study was a retrospective analysis of Komodo’s Healthcare Map, a large multi-payer US claims database. Pediatric and adult HSCT recipients were identified by Current Procedural Terminology, Healthcare Common Procedure Coding System, and ICD-10 procedure codes. HSCT recipients without a TMA code (controls) within 12 months following HSCT were matched 10:1 to patients with HSCT-TMA (cases). The index date was defined as the first claim for a TMA diagnosis (cases) or a date based on comparable post-HSCT time (controls), with follow-up until disenrollment or death. Results Between January 2016 and March 2024, 1,732 adults and 325 children who underwent HSCT and subsequently diagnosed with HSCT-TMA were matched to 18,691 and 2,335 controls, respectively. Over a mean follow-up of 18–29 months across different cohorts, both children and adults with HSCT-TMA experienced significantly higher complication rates compared to HSCT patients without HSCT-TMA including acute kidney injury, chronic kidney disease (CKD), hypertension, graft versus host disease (GVHD), sinusoidal obstructive syndrome, gastrointestinal (GI) bleeding, and blood stream infections (BSI; all p<0.05; Fig. 1).Absolute risk differences (RD) between cases and controls increased rapidly in the first 6 months post-index, then continued to rise for up to 3 years. Among children with HSCT-TMA, 3-year RDs for hypertension, CKD, and GI bleeding reached 33%, 13%, and 12%, respectively. Risk of BSI was 30% higher among HSCT-TMA patients vs. HSCT patients without HSCT-TMA at 6-month post-index and consistently elevated up to 3 years post-index. (Fig. 2).Adults with HSCT-TMA exhibited similarly elevated risks of clinical complications compared to HSCT patients without HSCT-TMA. Three-year RDs for hypertension, CKD, heart failure, ascites, and thrombosis were 20%, 15%, 12%, 10%, and 10%, respectively. Six-month RD for BSI reached 40% and remained stable up to 3-years post-index. (Fig. 3). Conclusions HSCT-TMA is associated with significant organ dysfunction and complications for up to 3 years post-diagnosis in survivors. Beyond the known increased rate of CKD, new findings show increased rates of heart failure, liver failure, BSI, and thrombosis in this untreated population that may not have been considered severe. These results suggest that HSCT-TMA may have lasting effects underscoring its substantial clinical burden and the importance of interventions to reduce its impact.

    2026TRANSPLANTATION AND CELLULAR THERAPY(2026)
    引用
    AI阅读
    加入学术空间
    立即登录,查看全部 1558 篇论文

    合作机构(100)

    葛兰素史克合作论文 77
    默克制药公司合作论文 51
    联合健康保险合作论文 45
    辉瑞合作论文 43
    安进合作论文 42
    明尼苏达大学合作论文 41
    诺华合作论文 41
    埃默里大学合作论文 31
    阿斯利康合作论文 28
    赛诺菲制药公司合作论文 27

    机构统计