Development, Deployment, and Implementation of a Machine Learning Surgical Case Length Prediction Model and Prospective Evaluation.

OBJECTIVE: To implement a machine learning model using only the restricted data available at case creation time to predict surgical case length for multiple services at different locations. BACKGROUND: The operating room is one of the most expensive resources in a health system, estimated to cost $22 to $133 per minute and generate about 40% of hospital revenue. Accurate prediction of surgical case length is necessary for efficient scheduling and cost-effective utilization of the operating room and other resources. METHODS: We introduced a similarity cascade to capture the complexity of cases and surgeon influence on the case length and incorporated that into a gradient-boosting machine learning model. The model loss function was customized to improve the balance between over- and under-prediction of the case length. A production pipeline was created to seamlessly deploy and implement the model across our institution. RESULTS: The prospective analysis showed that the model output was gradually adopted by the schedulers and outperformed the scheduler-predicted case length from August to December 2022. In 33,815 surgical cases across outpatient and inpatient platforms, the operational implementation predicted 11.2% fewer underpredicted cases and 5.9% more cases within 20% of the actual case length compared with the schedulers and only overpredicted 5.3% more. The model assisted schedulers to predict 3.4% more cases within 20% of the actual case length and 4.3% fewer underpredicted cases. CONCLUSIONS: We created a unique framework that is being leveraged every day to predict surgical case length more accurately at case posting time and could be potentially utilized to deploy future machine learning models.

Development and validation of a REcurrent Liver cAncer Prediction ScorE (RELAPSE) following liver transplantation in patients with hepatocellular carcinoma: Analysis of the US Multicenter HCC Transplant Consortium.

HCC recurrence following liver transplantation (LT) is highly morbid and occurs despite strict patient selection criteria. Individualized prediction of post-LT HCC recurrence risk remains an important need. Clinico-radiologic and pathologic data of 4981 patients with HCC undergoing LT from the US Multicenter HCC Transplant Consortium (UMHTC) were analyzed to develop a REcurrent Liver cAncer Prediction ScorE (RELAPSE). Multivariable Fine and Gray competing risk analysis and machine learning algorithms (Random Survival Forest and Classification and Regression Tree models) identified variables to model HCC recurrence. RELAPSE was externally validated in 1160 HCC LT recipients from the European Hepatocellular Cancer Liver Transplant study group. Of 4981 UMHTC patients with HCC undergoing LT, 71.9% were within Milan criteria, 16.1% were initially beyond Milan criteria with 9.4% downstaged before LT, and 12.0% had incidental HCC on explant pathology. Overall and recurrence-free survival at 1, 3, and 5 years was 89.7%, 78.6%, and 69.8% and 86.8%, 74.9%, and 66.7%, respectively, with a 5-year incidence of HCC recurrence of 12.5% (median 16 months) and non-HCC mortality of 20.8%. A multivariable model identified maximum alpha-fetoprotein (HR = 1.35 per-log SD, 95% CI,1.22-1.50, p < 0.001), neutrophil-lymphocyte ratio (HR = 1.16 per-log SD, 95% CI,1.04-1.28, p < 0.006), pathologic maximum tumor diameter (HR = 1.53 per-log SD, 95% CI, 1.35-1.73, p < 0.001), microvascular (HR = 2.37, 95%-CI, 1.87-2.99, p < 0.001) and macrovascular (HR = 3.38, 95% CI, 2.41-4.75, p < 0.001) invasion, and tumor differentiation (moderate HR = 1.75, 95% CI, 1.29-2.37, p < 0.001; poor HR = 2.62, 95% CI, 1.54-3.32, p < 0.001) as independent variables predicting post-LT HCC recurrence (C-statistic = 0.78). Machine learning algorithms incorporating additional covariates improved prediction of recurrence (Random Survival Forest C-statistic = 0.81). Despite significant differences in European Hepatocellular Cancer Liver Transplant recipient radiologic, treatment, and pathologic characteristics, external validation of RELAPSE demonstrated consistent 2- and 5-year recurrence risk discrimination (AUCs 0.77 and 0.75, respectively). We developed and externally validated a RELAPSE score that accurately discriminates post-LT HCC recurrence risk and may allow for individualized post-LT surveillance, immunosuppression modification, and selection of high-risk patients for adjuvant therapies.

Estimation of Surgery Durations Using Machine Learning Methods-A Cross-Country Multi-Site Collaborative Study.

The scheduling of operating room (OR) slots requires the accurate prediction of surgery duration. We evaluated the performance of existing Moving Average (MA) based estimates with novel machine learning (ML)-based models of surgery durations across two sites in the US and Singapore. We used the Duke Protected Analytics Computing Environment (PACE) to facilitate data-sharing and big data analytics across the US and Singapore. Data from all colorectal surgery patients between 1 January 2012 and 31 December 2017 in Singapore and, 1 January 2015 to 31 December 2019 in the US were used, and 7585 cases and 3597 single and multiple procedure cases from Singapore and US were included. The ML models were based on categorical gradient boosting (CatBoost) models trained on common data fields shared by both institutions. The procedure codes were based on the Table of Surgical Procedure (TOSP) (Singapore) and the Current Procedural Terminology (CPT) codes (US). The two types of codes were mapped by surgical experts. The CPT codes were then transformed into the relative value unit (RVU). The ML models outperformed the baseline MA models. The MA, scheduled durations and procedure codes were found to have higher loadings as compared to surgeon factors. We further demonstrated the use of the Duke PACE in facilitating data-sharing and big data analytics.

Social determinants of health data in solid organ transplantation: National data sources and future directions.

Health equity research in transplantation has largely relied on national data sources, yet the availability of social determinants of health (SDOH) data varies widely among these sources. We sought to characterize the extent to which national data sources contain SDOH data applicable to end-stage organ disease (ESOD) and transplant patients. We reviewed 10 active national data sources based in the United States. For each data source, we examined patient inclusion criteria and explored strengths and limitations regarding SDOH data, using the National Institutes of Health PhenX toolkit of SDOH as a data collection instrument. Of the 28 SDOH variables reviewed, eight-core demographic variables were included in ≥80% of the data sources, and seven variables that described elements of social status ranged between 30 and 60% inclusion. Variables regarding identity, healthcare access, and social need were poorly represented (≤20%) across the data sources, and five of these variables were included in none of the data sources. The results of our review highlight the need for improved SDOH data collection systems in ESOD and transplant patients via: enhanced inter-registry collaboration, incorporation of standardized SDOH variables into existing data sources, and transplant center and consortium-based investigation and innovation.

CPT to RVU conversion improves model performance in the prediction of surgical case length.

Methods used to predict surgical case time often rely upon the current procedural terminology (CPT) code as a nominal variable to train machine-learned models, however this limits the ability of the model to incorporate new procedures and adds complexity as the number of unique procedures increases. The relative value unit (RVU, a consensus-derived billing indicator) can serve as a proxy for procedure workload and could replace the CPT code as a primary feature for models that predict surgical case length. Using 11,696 surgical cases from Duke University Health System electronic health records data, we compared boosted decision tree models that predict individual case length, changing the method by which the model coded procedure type; CPT, RVU, and CPT-RVU combined. Performance of each model was assessed by inference time, MAE, and RMSE compared to the actual case length on a test set. Models were compared to each other and to the manual scheduler method that currently exists. RMSE for the RVU model (60.8 min) was similar to the CPT model (61.9 min), both of which were lower than scheduler (90.2 min). 65.2% of our RVU model's predictions (compared to 43.2% from the current human scheduler method) fell within 20% of actual case time. Using RVUs reduced model prediction time by ninefold and reduced the number of training features from 485 to 44. Replacing pre-operative CPT codes with RVUs maintains model performance while decreasing overall model complexity in the prediction of surgical case length.

Exploiting codon usage identifies intensity-specific modifiers of Ras/MAPK signaling in vivo.

Signal transduction pathways are intricately fine-tuned to accomplish diverse biological processes. An example is the conserved Ras/mitogen-activated-protein-kinase (MAPK) pathway, which exhibits context-dependent signaling output dynamics and regulation. Here, by altering codon usage as a novel platform to control signaling output, we screened the Drosophila genome for modifiers specific to either weak or strong Ras-driven eye phenotypes. Our screen enriched for regions of the genome not previously connected with Ras phenotypic modification. We mapped the underlying gene from one modifier to the ribosomal gene RpS21. In multiple contexts, we show that RpS21 preferentially influences weak Ras/MAPK signaling outputs. These data show that codon usage manipulation can identify new, output-specific signaling regulators, and identify RpS21 as an in vivo Ras/MAPK phenotypic regulator.

Wnt signaling suppresses MAPK-driven proliferation of intestinal stem cells.

Intestinal homeostasis depends on a slowly proliferating stem cell compartment in crypt cells, followed by rapid proliferation of committed progenitor cells in the transit amplifying (TA) compartment. The balance between proliferation and differentiation in intestinal stem cells (ISCs) is regulated by Wnt/ß-catenin signaling, although the mechanism remains unclear. We previously targeted PORCN, an enzyme essential for all Wnt secretion, and demonstrated that stromal production of Wnts was required for intestinal homeostasis. Here, a PORCN inhibitor was used to acutely suppress Wnt signaling. Unexpectedly, the treatment induced an initial burst of proliferation in the stem cell compartment of the small intestine, due to conversion of ISCs into TA cells with a loss of intrinsic ISC self-renewal. This process involved MAPK pathway activation, as the proliferating cells in the base of the intestinal crypt contained phosphorylated ERK1/2, and a MEK inhibitor attenuated the proliferation of ISCs and their differentiation into TA cells. These findings suggest a role for Wnt signaling in suppressing the MAPK pathway at the crypt base to maintain a pool of ISCs. The interaction between Wnt and MAPK pathways in vivo has potential therapeutic applications in cancer and regenerative medicine.

Engineering plasmonic nanorod arrays for colon cancer marker detection.

Engineering plasmonic nanomaterials or nanostructures towards ultrasensitive biosensing for disease markers or pathogens is of high importance. Here we demonstrate a systematic approach to tailor effective plasmonic nanorod arrays by combining both comprehensive numerical discrete dipole approximations (DDA) simulation and transmission spectroscopy experiments. The results indicate that 200×50 nm nanorod arrays with 300×500 nm period provide the highest figure of merit (FOM) of 2.4 and a sensitivity of 310 nm/RIU. Furthermore, we demonstrate the use of nanorod arrays for the detection of single nucleotide polymorphism in codon 12 of the K-ras gene that are frequently occurring in early stages of colon cancer, with a sensitivity down to 10 nM in the presence of 100-fold higher concentration of the homozygous genotypes. Our work shows significant potential of nanorod arrays towards point-of-care applications in diagnosis and clinical studies.

Stroma provides an intestinal stem cell niche in the absence of epithelial Wnts.

Wnt/ß-catenin signaling supports intestinal homeostasis by regulating proliferation in the crypt. Multiple Wnts are expressed in Paneth cells as well as other intestinal epithelial and stromal cells. Ex vivo, Wnts secreted by Paneth cells can support intestinal stem cells when Wnt signaling is enhanced with supplemental R-Spondin 1 (RSPO1). However, in vivo, the source of Wnts in the stem cell niche is less clear. Genetic ablation of Porcn, an endoplasmic reticulum resident O-acyltransferase that is essential for the secretion and activity of all vertebrate Wnts, confirmed the role of intestinal epithelial Wnts in ex vivo culture. Unexpectedly, mice lacking epithelial Wnt activity (Porcn(Del)/Villin-Cre mice) had normal intestinal proliferation and differentiation, as well as successful regeneration after radiation injury, indicating that epithelial Wnts are dispensable for these processes. Consistent with a key role for stroma in the crypt niche, intestinal stromal cells endogenously expressing Wnts and Rspo3 support the growth of Porcn(Del) organoids ex vivo without RSPO1 supplementation. Conversely, increasing pharmacologic PORCN inhibition, affecting both stroma and epithelium, reduced Lgr5 intestinal stem cells, inhibited recovery from radiation injury, and at the highest dose fully blocked intestinal proliferation. We conclude that epithelial Wnts are dispensable and that stromal production of Wnts can fully support normal murine intestinal homeostasis.