Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Desiderata for delivering NLP to accelerate healthcare AI advancement and a Mayo Clinic NLP-as-a-service implementation.

ABSTRACT: Data is foundational to high-quality artificial intelligence (AI). Given that a substantial amount of clinically relevant information is embedded in unstructured data, natural language processing (NLP) plays an essential role in extracting valuable information that can benefit decision making, administration reporting, and research. Here, we share several desiderata pertaining to development and usage of NLP systems, derived from two decades of experience implementing clinical NLP at the Mayo Clinic, to inform the healthcare AI community. Using a framework, we developed as an example implementation, the desiderata emphasize the importance of a user-friendly platform, efficient collection of domain expert inputs, seamless integration with clinical data, and a highly scalable computing infrastructure.

SUBMITTER: Wen A

PROVIDER: S-EPMC6917754 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

Desiderata for delivering NLP to accelerate healthcare AI advancement and a Mayo Clinic NLP-as-a-service implementation.

Wen Andrew A Fu Sunyang S Moon Sungrim S El Wazir Mohamed M Rosenbaum Andrew A Kaggal Vinod C VC Liu Sijia S Sohn Sunghwan S Liu Hongfang H Fan Jungwei J

NPJ digital medicine 20191217

Data is foundational to high-quality artificial intelligence (AI). Given that a substantial amount of clinically relevant information is embedded in unstructured data, natural language processing (NLP) plays an essential role in extracting valuable information that can benefit decision making, administration reporting, and research. Here, we share several desiderata pertaining to development and usage of NLP systems, derived from two decades of experience implementing clinical NLP at the Mayo Cl ...[more]

PMID: 31872069

Similar Datasets

Mayo Clinic Zebrafish Facility Overview.

Project description:The zebrafish (Danio rerio) is a premier nonmammalian vertebrate model organism. This small aquatic fish is utilized in multiple disciplines in the Mayo Clinic community and by many laboratories around the world because of its biological similarity to humans, its advanced molecular genetics, the elucidation of its genome sequence, and the ever-expanding and outstanding new biological tools now available to the zebrafish researcher. The Mayo Clinic Zebrafish Facility (MCZF) houses ?2,000 tanks annotated using an in-house, Internet cloud-based bar-coding system tied to our established zfishbook.org web infrastructure. Paramecia are the primary food source for larval fish rearing, using a simplified culture protocol described herein. The MCZF supports the specific ongoing research in a variety of laboratories, while also serving as a local hub for new scientists as they learn to tap into the potential of this model system for understanding normal development, disease, and as models of health.

| S-EPMC4931757 | biostudies-literature

Expression data from Mayo Clinic Pancreatic Tumor and Normal samples

2009-06-10 | GSE16515 | GEO

Mayo Clinic

Project description:Organisation EGAO00000000262

| EGAO00000000262 | EGA

Transcription profiling of human pancreatic tumor and normal samples from the Mayo clinic

Project description:We used microarrays to identify the expression differences of FKBP5 gene between the pancreatic tumor and normal samples. On average normal samples had more FKBP5 expression compared to tumor samples .Experiment Overall Design: This experiment consists of 36 tumor samples and 16 normal samples; a total of 52 samples. 16 samples consist of both tumor and normal expression data, whereas 20 samples consist of only tumor data.

2009-06-23 | E-GEOD-16515 | biostudies-arrayexpress

The Return of Actionable Variants Empirical (RAVE) Study, a Mayo Clinic Genomic Medicine Implementation Study: Design and Initial Results.

Project description:ObjectivesTo identify clinically actionable genetic variants from targeted sequencing of 68 disease-related genes, estimate their penetrance, and assess the impact of disclosing results to participants and providers.Patients and methodsThe Return of Actionable Variants Empirical (RAVE) Study investigates outcomes following the return of pathogenic/likely pathogenic (P/LP) variants in 68 disease-related genes. The study was initiated in December 2016 and is ongoing. Targeted sequencing was performed in 2533 individuals with hyperlipidemia or colon polyps. The electronic health records (EHRs) of participants carrying P/LP variants in 36 cardiovascular disease (CVD) genes were manually reviewed to ascertain the presence of relevant traits. Clinical outcomes, health care utilization, family communication, and ethical and psychosocial implications of disclosure of genomic results are being assessed by surveys, telephone interviews, and EHR review.ResultsOf 29,208 variants in the 68 genes, 1915 were rare (frequency <1%) and putatively functional, and 102 of these (60 in 36 CVD genes) were labeled P/LP based on the American College of Medical Genetics and Genomics framework. Manual review of the EHRs of participants (n=73 with P/LP variants in CVD genes) revealed that 33 had the expected trait(s); however, only 6 of 45 participants with non-familial hypercholesterolemia (FH) P/LP variants had the expected traits.ConclusionExpected traits were present in 13% of participants with P/LP variants in non-FH CVD genes, suggesting low penetrance; this estimate may change with additional testing performed as part of the clinical evaluation. Ongoing analyses of the RAVE Study will inform best practices for genomic medicine.

| S-EPMC6652203 | biostudies-literature

Quality improvement and healthcare: The Mayo Clinic quality Academy experience.

Project description: Not available

| S-EPMC7306605 | biostudies-literature

A 2030 Vision for the Mayo Clinic Department of Medicine.

Project description: Not available

| S-EPMC9247625 | biostudies-literature

Optimizing Door-to-Groin Puncture Time: The Mayo Clinic Experience.

Project description:ObjectivesTo provide a better understanding of methods that can be used to improve patient outcomes by reducing the door-to-groin puncture (DTP) time and present the results of a stroke quality improvement project (QIP) conducted by Mayo Clinic Arizona's stroke center.MethodsWe conducted a systematic literature search of Ovid MEDLINE(R), Ovid EMBASE, Scopus, and Web of Science for studies that evaluated DTP time reduction strategies. Those determined eligible for the purpose of this analysis were assessed for quality. The strategies for DTP time reduction were categorized on the basis of modified Target: Stroke Phase III recommendations and analyzed using a meta-analysis. The Mayo Clinic QIP implemented a single-call activation system to reduce DTP times by decreasing the time from neurosurgery notification to case start.ResultsFourteen studies were selected for the analysis, consisting of 2277 patients with acute ischemic stroke secondary to large-vessel occlusions. After intervention, all the studies showed a reduction in the DTP time, with the pooled DTP improvement being the standardized mean difference (1.37; 95% confidence interval, 1.20-1.93; τ2=1.09; P<.001). The Mayo Clinic QIP similarly displayed a DTP time reduction, with the DTP time dropping from 125.1 to 82.5 minutes after strategy implementation.ConclusionComputed tomography flow modifications produced the largest and most consistent reduction in the DTP time. However, the reduction in the DTP time across all the studies suggests that any systematic protocol aimed at reducing the DTP time can produce a beneficial effect. The relative novelty of mechanical thrombectomy and the consequential lack of research call for future investigation into the efficacy of varying DTP time reduction strategies.

| S-EPMC9253412 | biostudies-literature

Deep Brain Stimulator Device Infection: The Mayo Clinic Rochester Experience.

Project description:BackgroundDeep brain stimulator (DBS)-related infection is a recognized complication that may significantly alter the course of DBS therapy. We describe the Mayo Clinic Rochester experience with DBS-related infections.MethodsThis was a retrospective study of all adults (≥18 years old) who underwent DBS-related procedures between 2000 and 2020 at the Mayo Clinic Rochester.ResultsThere were 1087 patients who underwent 1896 procedures. Infection occurred in 57/1112 (5%) primary DBS implantations and 16/784 (2%) revision surgeries. The median time to infection (interquartile range) was 2.1 (0.9-6.9) months. The odds of infection were higher with longer operative length (P = .002), higher body mass index (BMI; P = .006), male sex (P = .041), and diabetes mellitus (P = .002). The association between infection and higher BMI (P = .002), male sex (P = .016), and diabetes mellitus (P = .003) remained significant in a subgroup analysis of primary implantations but not revision surgeries. Infection was superficial in 17 (23%) and deep in 56 (77%) cases. Commonly identified pathogens were Staphylococcus aureus (65%), coagulase-negative staphylococci (43%), and Cutibacterium acnes (45%). Three device management approaches were identified: 39 (53%) had complete device explantation, 20 (27%) had surgical intervention with device retention, and 14 (19%) had medical management alone. Treatment failure occurred in 16 (23%) patients. Time-to-event analysis showed fewer treatment failures with complete device explantation (P = .015). Only 1 individual had complications with brain abscess at failure.ConclusionsPrimary DBS implantations had higher rates of infection compared with revision surgeries. Complete device explantation was favored for deep infections. However, device salvage was commonly attempted and is a reasonable approach in select cases given the low rate of complications.

| S-EPMC9830487 | biostudies-literature

One thousand patients with primary myelofibrosis: the mayo clinic experience.

Project description:ObjectiveTo share our decades of experience with primary myelofibrosis and underscore the importance of outcomes research studies in designing clinical trials and interpreting their results.Patients and methodsOne thousand consecutive patients with primary myelofibrosis seen at Mayo Clinic between November 4, 1977, and September 1, 2011, were considered. The International Prognostic Scoring System (IPSS), dynamic IPSS (DIPSS), and DIPSS-plus were applied for risk stratification. Separate analyses were included for patients seen at time of referral (N=1000), at initial diagnosis (N=340), and within or after 1 year of diagnosis (N=660).ResultsTo date, 592 deaths and 68 leukemic transformations have been documented. Parameters at initial diagnosis vs time of referral included median age (66 vs 65 years), male sex (61% vs 62%), red cell transfusion need (24% vs 38%), hemoglobin level less than 10 g/dL (38% vs 54%), platelet count less than 100 × 10(9)/L (18% vs 26%), leukocyte count more than 25 × 10(9)/L (13% vs 16%), marked splenomegaly (21% vs 31%), constitutional symptoms (29% vs 34%), and abnormal karyotype (31% vs 41%). Mutational frequencies were 61% for JAK2V617F, 8% for MPLW515, and 4% for IDH1/2. DIPSS-plus risk distributions at time of referral were 10% low, 15% intermediate-1, 37% intermediate-2, and 37% high. The corresponding median survivals were 17.5, 7.8, 3.6, and 1.8 years vs 20.0, 14.3, 5.3, and 1.7 years for patients younger than 60 years of age. Compared with both DIPSS and IPSS, DIPSS-plus showed better discrimination among risk groups. Five-year leukemic transformation rates were 6% and 21% in low- and high-risk patients, respectively.ConclusionThe current document should serve as a valuable resource for patients and physicians and provides context for the design and interpretation of clinical trials.

| S-EPMC3538387 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data