Prediction of intra-abdominal injury using natural language processing of electronic medical record data.

BACKGROUND: This study aimed to use natural language processing to predict the presence of intra-abdominal injury using unstructured data from electronic medical records. METHODS: This was a random-sample retrospective observational cohort study leveraging unstructured data from injured patients taken to one of 9 acute care hospitals in an integrated health system between 2015 and 2021. Patients with International Classification of Diseases External Cause of Morbidity codes were identified. History and physical, consult, progress, and radiology report text from the first 8 hours of care were abstracted. Annotator dyads independently annotated encounters' text files to establish ground truth regarding whether intra-abdominal injury occurred. Features were extracted from text using natural language processing techniques, bag of words, and principal component analysis. We tested logistic regression, random forests, and gradient boosting machine to determine accuracy, recall, and precision of natural language processing to predict intra-abdominal injury. RESULTS: A random sample of 7,000 patient encounters of 177,127 was annotated. Only 2,951 had sufficient information to determine whether an intra-abdominal injury was present. Among those, 84 (2.9%) had an intra-abdominal injury. The concordance between annotators was 0.989. Logistic regression of features identified with bag of words and principal component analysis had the best predictive ability, with an area under the receiver operating characteristic curve of 0.9, recall of 0.73, and precision of 0.17. Text features with greatest importance included "abdomen," "pelvis," "spleen," and "hematoma." CONCLUSION: Natural language processing could be a screening decision support tool, which, if paired with human clinical assessment, can maximize precision of intra-abdominal injury identification.

The formulation of aptamer-coated paclitaxel-polylactide nanoconjugates and their targeting to cancer cells.

Paclitaxel-polylactide (Ptxl-PLA) conjugate nanoparticles, termed as nanoconjugates (NCs), were prepared through Ptxl/(BDI)ZnN(TMS)(2) (BDI = 2-((2,6-diisopropylphenyl)-amido)-4-((2,6-diisopropylphenyl)-imino)-2-pentene)-mediated controlled polymerization of lactide (LA) followed by nanoprecipitation. Nanoprecipitation of Ptxl-PLA resulted in sub-100 nm NCs with monomodal particle distributions and low polydispersities. The sizes of Ptxl-PLA NCs could be precisely controlled by using appropriate water-miscible solvents and by controlling the concentration of Ptxl-PLA during nanoprecipitation. Co-precipitation of a mixture of PLA-PEG-PLA (PLA = 14 kDa; PEG = 5 kDa) and Ptxl-PLA in PBS resulted in NCs that could stay non-aggregated in PBS for an extended period of time. To develop solid formulations of NCs, we evaluated a series of lyoprotectants, aiming to identify candidates that could effectively reduce or eliminate NC aggregation during lyophilization. Albumin was found to be an excellent lyoprotectant for the preparation of NCs in solid form, allowing lyophilized NCs to be readily dispersed in PBS without noticeable aggregates. Aptamer-NCs bioconjugates were prepared and found to be able to effectively target prostate-specific membrane antigen in a cell-specific manner.

Controlled formulation of doxorubicin-polylactide nanoconjugates for cancer drug delivery.

We report here the synthesis of doxorubicin-polylactide conjugate via doxorubicin-initiated lactide polymerization.