Clinical evaluation of a machine learning-based dysphagia risk prediction tool

PURPOSE

The rise of digitization promotes the development of screening and decision support tools. We sought to validate the results from a machine learning based dysphagia risk prediction tool with clinical evaluation.

METHODS

149 inpatients in the ENT department were evaluated in real time by the risk prediction tool, as well as clinically over a 3-week period. Patients were classified by both as patients at risk/no risk.

RESULTS

The AUROC, reflecting the discrimination capability of the algorithm, was 0.97. The accuracy achieved 92.6% given an excellent specificity as well as sensitivity of 98% and 82.4% resp. Higher age, as well as male sex and the diagnosis of oropharyngeal malignancies were found more often in patients at risk of dysphagia.

CONCLUSION

The proposed dysphagia risk prediction tool proved to have an outstanding performance in discriminating risk from no risk patients in a prospective clinical setting. It is likely to be particularly useful in settings where there is a lower incidence of patients with dysphagia and less awareness among staff.

Machine Learning-Based Prediction of Malnutrition in Surgical In-Patients: A Validation Pilot Study

BACKGROUND

Malnutrition in hospitalised patients can lead to serious complications, worse patient outcomes and longer hospital stays. State-of-the-art screening methods rely on scores, which need additional manual assessments causing higher workload.

OBJECTIVES

The aim of this prospective study was to validate a machine learning (ML)-based approach for an automated prediction of malnutrition in hospitalised patients.

METHODS

For 159 surgical in-patients, an assessment of malnutrition by dieticians was compared to the ML-based prediction conducted in the evening of admission.

RESULTS

The model achieved an accuracy of 83.0% and an AUROC of 0.833 in the prospective validation cohort.

CONCLUSION

The results of this pilot study indicate that an automated malnutrition screening could replace manual screening tools in hospitals.

Long-term risk of recurrent cerebrovascular events after patent foramen ovale closure: Results from a real-world stroke cohort

INTRODUCTION

Patent foramen ovale (PFO)-closure is recommended for stroke prevention in selected patients with suspected PFO-associated stroke. However, studies on cerebrovascular event recurrence after PFO-closure are limited by relatively short follow-up periods and information on the underlying aetiology of recurrent events is scarce.

PATIENTS AND METHODS

All consecutive patients with a cerebral ischaemic event and PFO-closure at the University Hospital Graz were prospectively identified from 2004 to 2021. Indication for PFO-closure was based on a neurological-cardiological PFO board decision. Patients underwent standardized clinical and echocardiographic follow-up 6 months after PFO-closure. Recurrent cerebrovascular events were assessed via electronical health records.

RESULTS

PFO-closure was performed in 515 patients (median age: 49 years; Amplatzer PFO occluder: 42%). Over a median follow-up of 11 years (range: 2-18 years, 5141 total patient-years), recurrent ischaemic cerebrovascular events were observed in 34 patients (ischaemic stroke: n = 22, TIA: n = 12) and associated with age, hyperlipidaemia and smoking in multivariable analysis (p < 0.05 each). Large artery atherosclerosis and small vessel disease were the most frequent aetiologies of recurrent stroke/TIA (27% and 24% respectively), and only two events were related to atrial fibrillation (AF). Recurrent ischaemic cerebrovascular event rates and incident AF were comparable in patients treated with different PFO occluders (p > 0.1).

DISCUSSION AND CONCLUSION

In this long-term follow-up-study of patients with a cerebral ischaemic event who had received PFO-closure with different devices, rates of recurrent stroke/TIA were low and largely related to large artery atherosclerosis and small vessel disease. Thorough vascular risk factor control seems crucial for secondary stroke prevention in patients treated for PFO-related stroke.

A Transformer-Based Model Trained on Large Scale Claims Data for Prediction of Severe COVID-19 Disease Progression

In situations like the COVID-19 pandemic, healthcare systems are under enormous pressure as they can rapidly collapse under the burden of the crisis. Machine learning (ML) based risk models could lift the burden by identifying patients with a high risk of severe disease progression. Electronic Health Records (EHRs) provide crucial sources of information to develop these models because they rely on routinely collected healthcare data. However, EHR data is challenging for training ML models because it contains irregularly timestamped diagnosis, prescription, and procedure codes. For such data, transformer-based models are promising. We extended the previously published Med-BERT model by including age, sex, medications, quantitative clinical measures, and state information. After pre-training on approximately 988 million EHRs from 3.5 million patients, we developed models to predict Acute Respiratory Manifestations (ARM) risk using the medical history of 80,211 COVID-19 patients. Compared to Random Forests, XGBoost, and RETAIN, our transformer-based models more accurately forecast the risk of developing ARM after COVID-19 infection. We used Integrated Gradients and Bayesian networks to understand the link between the essential features of our model. Finally, we evaluated adapting our model to Austrian in-patient data. Our study highlights the promise of predictive transformer-based models for precision medicine.

Secondary Use of Clinical Problem List Entries for Neural Network-Based Disease Code Assignment

Clinical information systems have become large repositories for semi-structured and partly annotated electronic health record data, which have reached a critical mass that makes them interesting for supervised data-driven neural network approaches. We explored automated coding of 50 character long clinical problem list entries using the International Classification of Diseases (ICD-10) and evaluated three different types of network architectures on the top 100 ICD-10 three-digit codes. A fastText baseline reached a macro-averaged F1-score of 0.83, followed by a character-level LSTM with a macro-averaged F1-score of 0.84. The top performing approach used a downstreamed RoBERTa model with a custom language model, yielding a macro-averaged F1-score of 0.88. A neural network activation analysis together with an investigation of the false positives and false negatives unveiled inconsistent manual coding as a main limiting factor.

Machine Learning Based Risk Prediction for Major Adverse Cardiovascular Events for ELGA-Authorized Clinics1

BACKGROUND

Artificial Intelligence (AI) has had an important impact on many industries as well as the field of medical diagnostics. In healthcare, AI techniques such as case-based reasoning and data driven machine learning (ML) algorithms have been used to support decision-making processes for complex tasks. This is used to assist medical professionals in making clinical decisions. A way of supporting clinicians is providing predicted prognoses of various ML models.

OBJECTIVES

Training an ML model based on the data of a hospital and using it on another hospital have some challenges.

METHODS

In this research, we applied data analysis to discover required data filters on a hospital's EHR data for training a model for another hospital.

RESULTS

We applied experiments on real-world data of ELGA (Austrian health record system) and KAGes (a public healthcare provider of 20+ hospitals in Austria). In this scenario, we train the prediction model for ELGA- authorized health service providers using the KAGes data since we do not have access to the complete ELGA data.

CONCLUSION

Finally, we observed that filtering the data with both feature and value selection increases the classification performance of the prediction model, which is trained for another system.

Frail People in LABLand: Development of an Easy-to-Use Machine Learning Model to Identify Frail People in Hospitals Based on Laboratory Data

BACKGROUND

Frail individuals are very vulnerable to stressors, which often lead to adverse outcomes. To ensure an adequate therapy, a holistic diagnostic approach is needed which is provided in geriatric wards. It is important to identify frail individuals outside the geriatric ward as well to ensure that they also benefit from the holistic approach.

OBJECTIVES

The goal of this study was to develop a machine learning model to identify frail individuals in hospitals. The model should be applicable without additional effort, quickly and in many different places in the healthcare system.

METHODS

We used Gradient Boosting Decision Trees (GBDT) to predict a frailty target derived from a gold standard assessment. The used features were laboratory values, age and sex. We also identified the most important features.

RESULTS

The best GBDT achieved an AUROC of 0.696. The most important laboratory values are urea, creatinine, granulocytes, chloride and calcium.

CONCLUSION

The model performance is acceptable, but insufficient for clinical use. Additional laboratory values or the laboratory history could improve the performance.

A combined active shape and mean appearance model for the reconstruction of segmental bone loss

This study investigates the novel combination of an active shape and mean appearance model to estimate missing bone geometry and density distribution from sparse inputs simulating segmental bone loss of the femoral diaphysis. An active shape Gaussian Process Morphable model was trained on healthy right femurs of South African males to model shape. The density distribution was approximated based on the mean appearance of computed tomography images from the training set. Estimations of diaphyseal resections were obtained by probabilistic fitting of the active shape model to sparse inputs consisting of proximal and distal femoral data on computed tomography images. The resulting shape estimates of the diaphyseal resections were then used to map the mean appearance model to the patients' missing bone geometry, constructing density estimations. In this way, resected bone surfaces were estimated with an average error of 2.24 (0.5) mm. Density distributions were approximated within 87 (0.7) % of the intensity of the original target images before the simulated segmental bone loss. These results fall within the acceptable tolerances required for surgical planning and reconstruction of long bone defects.

Explaining Machine Learning Predictions of Decision Support Systems in Healthcare

Artificial Intelligence (AI) methods, which are often based on Machine Learning (ML) algorithms, are also applied in the healthcare domain to provide predictions to physicians and patients based on electronic health records (EHRs), such as history of laboratory values, applied procedures and diagnoses. The question about these predictions “Why Should I Trust You?” encapsulates the issue with ML black boxes. Therefore, explaining the reasons for these ML predictions to physicians and patients is crucial to allow them to decide whether the prediction is applicable or not. In this paper, we explained and evaluated two prediction explanation methods for healthcare professionals (physicians and nurses). We compared two model-agnostic explanation methods based on global feature importance and local feature importance. We evaluated the user trust and reliance (UTR) for the explanation results of each method in a user study based on real patients’ electronic health records (EHR) and the feedback of healthcare professionals. Based on the user study, we observed that both methods have strengths and weaknesses according to the patients’ data, especially based on the data size of the patient. When the amount of data is small, global feature importance is enough to use. However, when the patient’s data size is big, using a local feature importance method makes more sense. As future work, we will develop a hybrid explanation method (by combining these methods automatically with a smart setting) to obtain higher and more stable performance results in terms of user trust and reliance.

External Validation of a Machine Learning Based Delirium Prediction Software in Clinical Routine

BACKGROUND

Various machine learning (ML) models have been developed for the prediction of clinical outcomes, but there is missing evidence on their performance in clinical routine and external validation.

OBJECTIVES

Our aim was to deploy and prospectively evaluate an already developed delirium prediction software in clinical routine of an external hospital.

METHODS

We compared updated ML models of the software and models re-trained with the external hospital's data. The best models were deployed in clinical routine for one month, and risk predictions for all admitted patients were compared to the risk ratings of a senior physician. After using the software, clinicians completed a questionnaire assessing technology acceptance.

RESULTS

Re-trained models achieved a high discriminative performance (AUROC > 0.92). Compared to clinical risk ratings, the software achieved a sensitivity of 100.0% and a specificity of 90.6%. Usefulness, ease of use and output quality were rated positively by the users.

CONCLUSION

A ML based delirium prediction software achieved a high discriminative performance and high technology acceptance at an external hospital using re-trained ML models.

Development of an Architecture to Implement Machine Learning Based Risk Prediction in Clinical Routine: A Service-Oriented Approach

BACKGROUND

Patients at risk of developing a disease have to be identified at an early stage to enable prevention. One way of early detection is the use of machine learning based prediction models trained on electronic health records.

OBJECTIVES

The aim of this project was to develop a software solution to predict cardiovascular and nephrological events using machine learning models. In addition, a risk verification interface for health care professionals was established.

METHODS

In order to meet the requirements, different tools were analysed. Based on this, a software architecture was created, which was designed to be as modular as possible.

RESULTS

A software was realised that is able to automatically calculate and display risks using machine learning models. Furthermore, predictions can be verified via an interface adapted to the need of health care professionals, which shows data required for prediction.

CONCLUSION

Due to the modularised software architecture and the status-based calculation process, different technologies could be applied. This facilitates the installation of the software at multiple health care providers, for which adjustments need to be carried out at one part of the software only.

Machine Learning Based Risk Prediction for Major Adverse Cardiovascular Events

BACKGROUND

Patients with major adverse cardiovascular events (MACE) such as myocardial infarction or stroke suffer from frequent hospitalizations and have high mortality rates. By identifying patients at risk at an early stage, MACE can be prevented with the right interventions.

OBJECTIVES

The aim of this study was to develop machine learning-based models for the 5-year risk prediction of MACE.

METHODS

The data used for modelling included electronic medical records of more than 128,000 patients including 29,262 patients with MACE. A feature selection based on filter and embedded methods resulted in 826 features for modelling. Different machine learning methods were used for modelling on the training data.

RESULTS

A random forest model achieved the best calibration and discriminative performance on a separate test data set with an AUROC of 0.88.

CONCLUSION

The developed risk prediction models achieved an excellent performance in the test data. Future research is needed to determine the performance of these models and their clinical benefit in prospective settings.

Risk prediction of delirium in hospitalized patients using machine learning: An implementation and prospective evaluation study

OBJECTIVE

Machine learning models trained on electronic health records have achieved high prognostic accuracy in test datasets, but little is known about their embedding into clinical workflows. We implemented a random forest-based algorithm to identify hospitalized patients at high risk for delirium, and evaluated its performance in a clinical setting.

MATERIALS AND METHODS

Delirium was predicted at admission and recalculated on the evening of admission. The defined prediction outcome was a delirium coded for the recent hospital stay. During 7 months of prospective evaluation, 5530 predictions were analyzed. In addition, 119 predictions for internal medicine patients were compared with ratings of clinical experts in a blinded and nonblinded setting.

RESULTS

During clinical application, the algorithm achieved a sensitivity of 74.1% and a specificity of 82.2%. Discrimination on prospective data (area under the receiver-operating characteristic curve = 0.86) was as good as in the test dataset, but calibration was poor. The predictions correlated strongly with delirium risk perceived by experts in the blinded (r = 0.81) and nonblinded (r = 0.62) settings. A major advantage of our setting was the timely prediction without additional data entry.

DISCUSSION

The implemented machine learning algorithm achieved a stable performance predicting delirium in high agreement with expert ratings, but improvement of calibration is needed. Future research should evaluate the acceptance of implemented machine learning algorithms by health professionals.

CONCLUSIONS

Our study provides new insights into the implementation process of a machine learning algorithm into a clinical workflow and demonstrates its predictive power for delirium.

Technology Acceptance of a Machine Learning Algorithm Predicting Delirium in a Clinical Setting: a Mixed-Methods Study

Early identification of patients with life-threatening risks such as delirium is crucial in order to initiate preventive actions as quickly as possible. Despite intense research on machine learning for the prediction of clinical outcomes, the acceptance of the integration of such complex models in clinical routine remains unclear. The aim of this study was to evaluate user acceptance of an already implemented machine learning-based application predicting the risk of delirium for in-patients. We applied a mixed methods design to collect opinions and concerns from health care professionals including physicians and nurses who regularly used the application. The evaluation was framed by the Technology Acceptance Model assessing perceived ease of use, perceived usefulness, actual system use and output quality of the application. Questionnaire results from 47 nurses and physicians as well as qualitative results of four expert group meetings rated the overall usefulness of the delirium prediction positively. For healthcare professionals, the visualization and presented information was understandable, the application was easy to use and the additional information for delirium management was appreciated. The application did not increase their workload, but the actual system use was still low during the pilot study. Our study provides insights into the user acceptance of a machine learning-based application supporting delirium management in hospitals. In order to improve quality and safety in healthcare, computerized decision support should predict actionable events and be highly accepted by users.

Abstract PS13-11: Oral paclitaxel and encequidar (oPac+E) in the treatment of metastatic breast cancer (mBC): Management of gastrointestinal adverse events (GI AE). Study KX-ORAX-001

Background: There is a need for more effective and less toxic treatments for patients with mBC. Patients may prefer oral vs IV cytotoxic therapies to avoid frequent hospital visits. In addition, oral therapies allow frequent or metronomic dosing regimens which may alter the toxicity or activity profile of agents vs infrequent IV administration. oPac+E is oral paclitaxel combination with Encequidar, a specific, minimally absorbed, oral p-glycoprotein inhibitor that facilitates the absorption of oral paclitaxel. mBC patients who received oPac+E had significantly greater confirmed tumor response and longer survival with lower rates and severity of neuropathy but increased GI AE compared to IV paclitaxel (IVPac) (Study KX-ORAX-001 presented at SABCS, 2019, Abstract # GS6-01).

Methods: Study KX-ORAX-001 was a phase III, randomized, study in women with mBC for whom treatment with IVPac was recommended. Patients were randomized 2:1 to receive oPac+E or IVPac. Patients continued treatment until discontinuation due to progressive disease or toxicity. oPac 205 mg/m2 was given once daily for 3 days weekly. E 12.9 mg was given 1 hour before each dose of oPac. IVPac 175 mg/m2 was infused over 3 hours every 3 weeks. The primary endpoint was efficacy defined as tumor response confirmed by BICR at two consecutive evaluations. Key secondary endpoints included PFS, OS. Safety was monitored throughout the study.

Results: All IVPac patients received high-dose dexamethasone and antihistamine premedication, which have significant anti-emetic activity and may have received additional anti-emetic agents as needed. The protocol did not allow any prophylaxis for GI AE for oPac+E patients nor were they to receive predose corticosteroids, nor antihistamines.

The protocol was amended after approximately 30% of patients were enrolled to allow prophylactic anti-emetic medications for patients randomized to oPac+E. Patients were also given loperamide to take at home and were instructed to initiate loperamide with the onset of diarrhea. The rates of Grade ≥2, vomiting and diarrhea for patients treated with IVPac, the patients treated with oPac+E prior to after the amendment are summarized in the table below.

Prophylactic anti-emetic therapy and early use of loperamide markedly decreased the incidence of ≥Grade 2 vomiting and diarrhea although there was a greater incidence than IVPac.

The most frequently prescribed anti-emetic agents for oPac+E treated patients were ondansetron (54%), metoclopramide (21%), domperidone (4%) and aprepitant (3%). For patients randomized to IVPac, the most frequently prescribed agents were ondanesteron (59%), granisetron (24%), palenosetron (7%) and aprepitant (2%). Oral administration of the oral NK1 inhibitor aprepitant appeared to be associated with increased incidence of oral paclitaxel systemic toxicity, potentially due to inhibition of metabolism of oPac by cytochrome P450 3A4.

Conclusions: oPac+E was associated with greater efficacy in the treatment of mBC and lower rates and severity of peripheral neuropathy, but increased GI AE compared to IVPac 175mg/m2. GI AE in oPac+E treated patients can be managed by prophylactic use of anti-emetics, primarily 5-HT3 inhibitors and early intervention with the anti-diarrhea agent loperamide. The use of the oral NK1 inhibitor aprepitant in combination with oPac+E is not recommended.(NTC02594371)

IVPacoPac+E Pre-AmendmentoPac+E Post AmendmentGrade 2Grade 3Grade 4Grade 2Grade 3Grade 4Grade 2Grade 3Grade 4Vomiting4%1%0%24%7%0%7%4%0%Diarrhea7%1%0%27%9%0%16%3%0.5%

Citation Format: H S Rugo, G Umanzor, F J Barrios, R H Vasallo, M A Chivalan, S Bejarano, J R Ramirez, L Fein, R D Kowalyszyn, D L Cutler, D Kramer, J Goldfinch, H Wang, T Moore, R MF Kwan. Oral paclitaxel and encequidar (oPac+E) in the treatment of metastatic breast cancer (mBC): Management of gastrointestinal adverse events (GI AE). Study KX-ORAX-001 [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-11.

Abstract PS13-06: Lower rates of neuropathy with oral paclitaxel and encequidar (oPac+E) compared to IV paclitaxel (IVPac) in treatment of metastatic breast cancer (mBC): Study KX-ORAX-001

Background: Chemotherapy-induced peripheral neuropathy (CIPN) is a common dose-limiting toxicity associated with IVPac. Primarily sensory, CIPN is an often irreversible condition primarily affecting the hands and feet associated with pain, numbness, tingling, and sensitivity to cold and has a significant impact on quality of life and treatment tolerance. Risk of CIPN increases with age, dose intensity, cumulative dose, and preexisting conditions including diabetes.

Methods: Study KX-ORAX-001 was a phase III, randomized, international study in women with mBC for whom treatment with IVPac was recommended. Eligible patients were randomized 2:1 to receive oPac+E or IVPac. Patients continued treatment until discontinuation due to progressive disease or toxicity. oPac 205 mg/m2 was given once daily for 3 days weekly. E 12.9 mg was given 1 hour before each dose of oPac. IVPac 175 mg/m2 was infused over 3 hours every 3 weeks. The primary endpoint was efficacy defined as tumor response confirmed by BICR at two consecutive evaluations. Key secondary endpoints included PFS, OS. Safety was monitored throughout the study.

Results: A total of 402 mBC patients were enrolled, 265 randomized to oPac+E and 137 to IVPac (ITT population). 399 patients were treated and comprise the safety population. The confirmed response rate was significantly greater in the oPac+E group vs IVPac (35% vs 23%) for the ITT population. Median overall survival was (27.7 vs 16.7 months, ITT) at the time of the analysis. Long-term follow up for final determination of PFS and OS is ongoing.Incidence of neuropathy-related TEAEs were lower in patients receiving oPac+E vs IVPac: Overall (21% vs 64%; all grades), grade ≥3 (2% vs 15%). Cumulative risk for neuropathy with IVPac was &gt;50% by week 8 and was 83% at week 88. In contrast, the cumulative risk of neuropathy with oPac+E rose slowly and plateaued at 34% at week 88. Treatment discontinuations due to neuropathy occurred only in the IVPac arm (8%). Dose reductions due to neuropathy were reported in 8% of IVPac treated patients and in 2% of oPac+E treated patients. In agreement with the lower rates of peripheral neuropathy in patients treated with oPac+E, there was lower use of medications used for the treatment of neuropathic symptoms. Use of gabapentin or pregabalin was 12% for patients receiving oPac+E vs 40% for IVPac treated patients.

Conclusions: oPac+E was associated with greater efficacy in the treatment of patients with mBC and a lower incidence of neuropathy, slower onset and lesser severity of neuropathic events compared to IVPac 175mg/m2 administered every three weeks. Fewer patients receiving oPac+E required dose reduction due to neuropathy and no patients receiving oPac+E discontinued treatment due to neuropathy. Reduction in neuropathy may improve quality of life and allow longer administration of effective therapy while maintaining dose intensity.

Citation Format: H S Rugo, G Umanzor, F J Barrios, R H Vasallo, M A Chivalan, S Bejarano, J R Ramirez, L Fein, R D Kowalyszyn, D L Cutler, D Kramer, J Goldfinch, H Wang, T Moore, R MF Kwan. Lower rates of neuropathy with oral paclitaxel and encequidar (oPac+E) compared to IV paclitaxel (IVPac) in treatment of metastatic breast cancer (mBC): Study KX-ORAX-001 [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS13-06.

10th European immunogenicity platform open symposium on immunogenicity of biopharmaceuticals

Therapeutic proteins and emerging gene and cell-based therapies are attractive therapeutic tools for addressing unmet medical needs or when earlier conventional treatment approaches failed. However, the development of an immune response directed against therapeutic agents is a significant concern as it occurs in a substantial number of cases across products and indications. The specific anti-drug antibodies that develop can lead to safety adverse events as well as inhibition of drug activity or accelerated clearance, both phenomena resulting in loss of treatment efficacy. The European Immunogenicity Platform (EIP) is a meeting place for experts and newcomers to the immunogenicity field, designed to stimulate discussion amongst scientists across industry and academia, encourage interactions with regulatory agencies and share knowledge and the state-of-the-art of immunogenicity sciences with the broader scientific community. Here we report on the main topics covered during the EIP 10th Open Symposium on Immunogenicity of Biopharmaceuticals held in Lisbon, 26–27 February 2019, and the 1-d training course on practical and regulatory aspects of immunogenicity held ahead of the conference. These main topics included immunogenicity testing, clinical relevance of immunogenicity, immunogenicity prediction, regulatory aspects, tolerance induction as a mean to mitigate immunogenicity and immunogenicity in the context of gene therapy.

Multivariable Risk Prediction of Dysphagia in Hospitalized Patients Using Machine Learning

BACKGROUND

Dysphagia is a dysfunction of the swallowing act and is highly prevalent in acute post-stroke patients and patients with chronic neurological diseases. Dysphagia is associated with several potentially life threatening complications. Thus, an early identification and treatment could reduce morbidity and mortality rates.

OBJECTIVES

The aim of the study was to develop a multivariable model predicting the individual risk of dysphagia in hospitalized patients.

METHODS

We trained different machine learning algorithms on the electronic health records of over 33,000 patients.

RESULTS

The tree-based Random Forest Classifier and Adaboost Classifier algorithms achieved an area under the receiver operating characteristic curve of 0.94.

CONCLUSION

The developed models outperformed previously published models predicting dysphagia. In future, an implementation in the clinical workflow is needed to determine the clinical benefit.

A novel hybrid modeling approach for the evaluation of integrated care and economic outcome in heart failure treatment

Background

Demographic changes, increased life expectancy and the associated rise in chronic diseases pose challenges to public health care systems. Optimized treatment methods and integrated concepts of care are potential solutions to overcome increasing financial burdens and improve quality of care. In this context modeling is a powerful tool to evaluate potential benefits of different treatment procedures on health outcomes as well as health care budgets. This work presents a novel modeling approach for simulating different treatment procedures of heart failure patients based on extensive data sets from outpatient and inpatient care.

Methods

Our hybrid heart failure model is based on discrete event and agent based methodologies and facilitates the incorporation of different therapeutic procedures for outpatient and inpatient care on patient individual level. The state of health is modeled with the functional classification of the New York Heart Association (NYHA), strongly affecting discrete state transition probabilities alongside age and gender. Cooperation with Austrian health care and health insurance providers allowed the realization of a detailed model structure based on clinical data of more than 25,000 patients.

Results

Simulation results of conventional care and a telemonitoring program underline the unfavorable prognosis for heart failure and reveal the correlation of NYHA classes with health and economic outcomes. Average expenses for the treatment of NYHA class IV patients of €10,077 ± €165 were more than doubled compared to other classes. The selected use case of a telemonitoring program demonstrated potential cost savings within two years of application. NYHA classes II and III revealed most potential for additional treatment measures.

Conclusion

The presented model allows performing extensive simulations of established treatment procedures for heart failure patients and evaluating new holistic methods of care and innovative study designs. This approach offers health care providers a unique, adaptable and comprehensive tool for decision making in the complex and socioeconomically challenging field of cardiovascular diseases.

Evolution of femtosecond laser damage in a hafnia-silica multi-layer dielectric coating

To optimize optical coating materials, designs, and technologies for high damage resistance, understanding the growth of laser damage is of paramount importance. In this Letter, we show the evolution of femtosecond laser damage in a hafnia-silica (HfO₂/SiO₂) multilayer dielectric mirror coating. Depending on various spatial features of damaged sites, we identified several regimes of the laser-material interaction with varying laser fluence and incident number of pulses. A change in surface roughness has been observed only for a small number of pulses, and interestingly, a threshold number of pulses is found for nanocrack formation. We report the polarization-dependent orientation of nanocracks and their growth with an increasing number of pulses. The presented results demonstrate that the laser damage originates from the nanobumps and surface roughening, which then leads to the formation of nanocracks. The presented experimental results acknowledge the existing theoretical models in bulk dielectrics to explain the formation of nanostructures by interference of the incident laser with the scattering radiation from laser-induced inhomogeneities and growth of the field enhancement due to nanoplasma.

An Improvised Classification Model for Predicting Delirium

With the vast increase of digital healthcare data, there is an opportunity to mine the data for understanding inherent health patterns. Although machine-learning techniques demonstrated their applications in healthcare to answer several questions, there is still room for improvement in every aspect. In this paper, we are demonstrating a method that improves the performance of a delirium prediction model using random forest in combination with logistic regression.

Development of a Machine Learning Model Predicting an ICU Admission for Patients with Elective Surgery and Its Prospective Validation in Clinical Practice

Frequent utilization of the Intensive Care Unit (ICU) is associated with higher costs and decreased availability for patients who urgently need it. Common risk assessment tool, like the ASA score, lack objectivity and do account only for some influencing parameters. The aim of our study was (1) to develop a reliable machine learning model predicting ICU admission risk after elective surgery, and (2) to implement it in a clinical workflow. We used electronic medical records from more than 61,000 patients for modelling. A random forest model outperformed other methods with an area under the curve of 0.91 in the retrospective test set. In the prospective implementation, the model achieved a sensitivity of 73.3% and a specificity of 80.8%. Further research is essential to determine physicians' attitudes to machine learning models and assess the long term improvement of ICU management.

EHR problem list clustering for improved topic-space navigation

Background

The amount of patient-related information within clinical information systems accumulates over time, especially in cases where patients suffer from chronic diseases with many hospitalizations and consultations. The diagnosis or problem list is an important feature of the electronic health record, which provides a dynamic account of a patient’s current illness and past history. In the case of an Austrian hospital network, problem list entries are limited to fifty characters and are potentially linked to ICD-10. The requirement of producing ICD codes at each hospital stay, together with the length limitation of list items leads to highly redundant problem lists, which conflicts with the physicians’ need of getting a good overview of a patient in short time.

This paper investigates a method, by which problem list items can be semantically grouped, in order to allow for fast navigation through patient-related topic spaces.

Methods

We applied a minimal language-dependent preprocessing strategy and mapped problem list entries as tf-idf weighted character 3-grams into a numerical vector space. Based on this representation we used the unweighted pair group method with arithmetic mean (UPGMA) clustering algorithm with cosine distances and inferred an optimal boundary in order to form semantically consistent topic spaces, taking into consideration different levels of dimensionality reduction via latent semantic analysis (LSA).

Results

With the proposed clustering approach, evaluated via an intra- and inter-patient scenario in combination with a natural language pipeline, we achieved an average compression rate of 80% of the initial list items forming consistent semantic topic spaces with an F-measure greater than 0.80 in both cases. The average number of identified topics in the intra-patient case (μ_Intra = 78.4) was slightly lower than in the inter-patient case (μ_Inter = 83.4). LSA-based feature space reduction had no significant positive performance impact in our investigations.

Conclusions

The investigation presented here is centered on a data-driven solution to the known problem of information overload, which causes ineffective human-computer interactions at clinicians’ work places. This problem is addressed by navigable disease topic spaces where related items are grouped and the topics can be more easily accessed.

Reduced fetal growth velocities and the association with neonatal outcomes in appropriate-for-gestational-age neonates: a retrospective cohort study

Background

Fetal growth restriction is, despite advances in neonatal care and uptake of antenatal ultrasound scanning, still a major cause of perinatal morbidity. Neonates with birth weight > 10th percentile are assumed to be appropriate-for-gestational-age (AGA), although many are at increased risk of perinatal morbidity, because of undetected mild restriction of growth potential. We hypothesized that within AGA neonates, reduced fetal growth velocities are associated with adverse neonatal outcome.

Methods

A retrospective cohort study of singleton pregnancies, in the Maastricht University Medical Centre (MUMC) between 2010 and 2016. Women had two fetal biometry scans (18–22 weeks and 30–34 weeks of gestational age) and delivered a newborn with a birth weight between the 10th–80th percentile.

Differences in growth velocities of the abdominal circumference (AC), biparietal diameter (BPD), head circumference (HC) and femur length (FL) were compared between the suboptimal AGA (sAGA) (birth weight centiles 10–50) and optimal AGA (oAGA) (birth weight centiles 50–80) group. We assessed the association between velocities and neonatal outcomes.

Results

We included 934 singleton pregnancies. In the suboptimal AGA group, fetal growth velocities were lower (in mm/week): AC 10.72 ± 1.00 vs 11.23 ± 1.00 (p < .001), HC 10.50 ± 0.80 vs 10.68 ± 0.77 (p = 0.001), BPD 3.01 ± 0.28 vs 3.08 ± 0.27 (p < .0001) and FL 2.47 ± 0.21 vs 2.50 ± 0.22 (p = 0.014), compared to the optimal AGA group. Neonates with an adverse neonatal outcome had significantly lower growth velocities (in mm/week) of: AC 10.57 vs 10.94 (p = 0.034), HC 10.28 vs 10.59 (p = 0.003) and BPD 2.97 vs 3.04 (p = 0.043) compared to those with normal outcome. An inverse association was observed between the AC velocity and a composite adverse neonatal outcome (OR) = 0.667 (95%CI 0.507–0.879, p = 0.004), and between the AC velocity and neonates with NICU stay (OR) = 0.733 (95%CI 0.570–0.942, p = 0.015). Neonates with a birthweight lower than expected (based on the abdominal circumference at 20 weeks) had significantly more composite adverse neonatal outcomes 8.5% vs 5.0% (p = 0.047), NICU stays 9.6% vs 3.8% (p < .0001) and hospital stays 44.4% vs 35.6% (p = 0.006).

Conclusions

Appropriate-for-gestational-age neonates are a heterogeneous group with some showing suboptimal fetal growth. Abnormal fetal growth velocities, especially abdominal circumference velocity, are associated with adverse neonatal outcome and can potentially improve the detection of mild growth restriction when used in multivariate models.

Information Adapted Machine Learning Models for Prediction in Clinical Workflow

BACKGROUND

In a database of electronic health records, the amount of available information varies widely between patients. In a real-time prediction scenario, a machine learning model may receive limited information for some patients.

OBJECTIVES

Our aim was to evaluate the influence of missing data on real-time prediction of delirium, and detect changes in prediction performance when training separate models for patients with missing data.

METHODS

We compared a model trained specifically on data with missing values to the currently implemented model predicting delirium. Also, we simulated five test data sets with different amount of missing data and compared the prediction results to the prediction on complete data set when using the same model.

RESULTS

For patients with missing laboratory and nursing assessment data, a model trained especially for this scenario performed significantly better than the implemented model. The combination of procedure data and demographic data achieved the closest results to a prediction with a complete data set.

CONCLUSION

An ongoing evaluation of real-time prediction is indispensable. Additional models adapted to the information available might improve prediction performance.

Is Regular Re-Training of a Predictive Delirium Model Necessary After Deployment in Routine Care?

Adoption of electronic medical records in hospitals generates a large amount of data. Health care professionals can easily lose their sight on the important insights of the patients' clinical and medical history. Although machine learning algorithms have already proved their significance in healthcare research, remains a challenge translation and dissemination of fully automated prediction algorithms from research to decision support at the point of care. In this paper, we address the effect of changes in the characteristics of data over time on the performance of deployed models for the use case of predicting delirium in hospitalised patients. We have analysed the stability of models trained with subsets of data from one single year (2012, 2013...2016, respectively), and tested the models with data from 2017. Our results show that in the case of delirium prediction, the models were stable over time, indicating that re-training the models is not necessary e.g. once per year might be more than sufficient.

Effect of Nursing Assessment on Predictive Delirium Models in Hospitalised Patients

Delirium is an acute neuropsychiatric syndrome which is common in elderly patients during their hospitalisation and is associated with an increased mortality and morbidity. Since delirium is a) often underdiagnosed and b) preventable if early signs are detected,igh expectations are set in delirium risk assessment during hospital admission. In our latest studies, we showed that delirium prediction using machine learning algorithms is possible based on the patients' health history. The aim of this study is to compare the influence of nursing assessment data on prediction models with clinical and demographic data. We approached the problem by a) comparing the performance of predictive models including nursing data with models based on clinical and demographic data only and b) analysing the feature importance of all available features. From our results we concluded that nursing assessment data can improve the performance of delirium prediction models better than demographic, laboratory, diagnosis, procedures, and previous transfers' data alone.

Cleansing and Imputation of Body Mass Index Data and Its Impact on a Machine Learning Based Prediction Model

BACKGROUND

A challenge of using electronic health records for secondary analyses is data quality. Body mass index (BMI) is an important predictor for various diseases but often not documented properly.

OBJECTIVES

The aim of our study is to perform data cleansing on BMI values and to find the best method for an imputation of missing values in order to increase data quality. Further, we want to assess the effect of changes in data quality on the performance of a prediction model based on machine learning.

METHODS

After data cleansing on BMI data, we compared machine learning methods and statistical methods in their accuracy of imputed values using the root mean square error. In a second step, we used three variations of BMI data as a training set for a model predicting the occurrence of delirium.

RESULTS

Neural network and linear regression models performed best for imputation. There were no changes in model performance for different BMI input data.

CONCLUSION

Although data quality issues may lead to biases, it does not always affect performance of secondary analyses.

EHR Text Categorization for Enhanced Patient-Based Document Navigation

Patients with multiple disorders usually have long diagnosis lists, constitute by ICD-10 codes together with individual free-text descriptions. These text snippets are produced by overwriting standardized ICD-Code topics by the physicians at the point of care. They provide highly compact expert descriptions within a 50-character long text field frequently not assigned to a specific ICD-10 code. The high redundancy of these lists would benefit from content-based categorization within different hospital-based application scenarios. This work demonstrates how to accurately group diagnosis lists via a combination of natural language processing and hierarchical clustering with an overall F-measure value of 0.87. In addition, it compresses the initial diagnosis list up to 89%. The manuscript discusses pitfall and challenges as well as the potential of a large-scale approach for tackling this problem.

On the Representation of Machine Learning Results for Delirium Prediction in a Hospital Information System in Routine Care

Digitalisation of health care for the purpose of medical documentation lead to huge amounts of data, hence having an opportunity to derive knowledge and associations of different attributes recorded. Many health care events can be prevented when identified. Machine learning algorithms could identify such events but there is ambiguity in understanding the suggestions especially in clinical setup. In this paper we are presenting how we explain the decision based on random forest to health care professionals in the course of the project predicting delirium during hospitalisation on the day of admission.

Evaluating the Impact of Incorrect Diabetes Coding on the Performance of Multivariable Prediction Models

The use of electronic health records for risk prediction models requires a sufficient quality of input data to ensure patient safety. The aim of our study was to evaluate the influence of incorrect administrative diabetes coding on the performance of a risk prediction model for delirium, as diabetes is known to be one of the most relevant variables for delirium prediction. We used four data sets varying in their correctness and completeness of diabetes coding as input for different machine learning algorithms. Although there was a higher prevalence of diabetes in delirium patients, the model performance parameters did not vary between the data sets. Hence, there was no significant impact of incorrect diabetes coding on the performance for our model predicting delirium.

Development and Validation of a Multivariable Prediction Model for the Occurrence of Delirium in Hospitalized Gerontopsychiatry and Internal Medicine Patients

Delirium is an acute confusion condition, which is common in elderly and often misdiagnosed in hospitalized patients. Early identification and prevention of delirium could reduce morbidity and mortality rates in those affected and reduce hospitalization costs. We have developed and validated a multivariate prediction model that predicts delirium and gives an early warning to physicians. A large set of patient electronic medical records have been used in developing the models. Classical learning algorithms have been used to develop the models and compared the results. Excellent results were obtained with the feature set and parameter settings attaining accuracy of 84%.

En Bloc Liver Kidney Transplantation Using Donor Splenic Artery as Inflow to the Kidney: Report of Two Cases

The number of simultaneous liver-kidney transplants has been increasing. This surgery is associated with an increased risk of complications, longer duration of surgery and longer ischemia time for the renal allograft. Two patients listed for liver-kidney transplant at our center underwent en bloc combined liver-kidney transplantation using donor splenic artery as inflow. Patient 1 previously underwent cardiac catheterization that was complicated by a bleeding pseudoaneurysm of the right external iliac artery that required endovascular stenting of the external iliac artery and embolization of the inferior epigastric artery. Patient 2 was on vasopressor support and continuous renal replacement therapy at the time of transplant. In this paper, we described a novel technique of en bloc liver-kidney transplant with simultaneous reperfusion of both allografts using the donor splenic artery for renal inflow. This technique is useful for decreasing cold ischemia time and total operative time by simultaneous reperfusion of both allografts. It is a useful technical variant that can be used in patients with severe disease of the iliac arteries.

Standardizing terms, definitions and concepts for describing and interpreting unwanted immunogenicity of biopharmaceuticals: recommendations of the Innovative Medicines Initiative ABIRISK consortium

Biopharmaceuticals (BPs) represent a rapidly growing class of approved and investigational drug therapies that is contributing significantly to advancing treatment in multiple disease areas, including inflammatory and autoimmune diseases, genetic deficiencies and cancer. Unfortunately, unwanted immunogenic responses to BPs, in particular those affecting clinical safety or efficacy, remain among the most common negative effects associated with this important class of drugs. To manage and reduce risk of unwanted immunogenicity, diverse communities of clinicians, pharmaceutical industry and academic scientists are involved in: interpretation and management of clinical and biological outcomes of BP immunogenicity, improvement of methods for describing, predicting and mitigating immunogenicity risk and elucidation of underlying causes. Collaboration and alignment of efforts across these communities is made difficult due to lack of agreement on concepts, practices and standardized terms and definitions related to immunogenicity. The Innovative Medicines Initiative (IMI; http://www.imi-europe.org), ABIRISK consortium [Anti-Biopharmaceutical (BP) Immunization Prediction and Clinical Relevance to Reduce the Risk; http://www.abirisk.eu] was formed by leading clinicians, academic scientists and EFPIA (European Federation of Pharmaceutical Industries and Associations) members to elucidate underlying causes, improve methods for immunogenicity prediction and mitigation and establish common definitions around terms and concepts related to immunogenicity. These efforts are expected to facilitate broader collaborations and lead to new guidelines for managing immunogenicity. To support alignment, an overview of concepts behind the set of key terms and definitions adopted to date by ABIRISK is provided herein along with a link to access and download the ABIRISK terms and definitions and provide comments (http://www.abirisk.eu/index_t_and_d.asp).

A pro-apoptotic function of iASPP by stabilizing p300 and CBP through inhibition of BRMS1 E3 ubiquitin ligase activity

The p53 family and its cofactors are potent inducers of apoptosis and form a barrier to cancer. Here, we investigated the impact of the supposedly inhibitory member of the apoptosis-stimulating protein of p53, iASPP, on the activity of the p53 homolog TAp73, and its cofactors p300 and CBP. We found that iASPP interacted with and stabilized the histone acetyltransferase p300 and its homolog CBP upon cisplatin treatment. Vice versa, iASPP depletion by shRNA resulted in decreased amounts of p300 and CBP, impaired binding of p300 and TAp73 to target site promoters, reduced induction of pro-apoptotic TAp73 target genes, and impaired apoptosis. Mechanistically, we observed that the p300-regulatory E3 ubiquitin ligase BRMS1 could rescue the degradation of p300 and CBP in cisplatin-treated, iASPP-depleted cells. This argues that iASPP stabilizes p300 and CBP by interfering with their BRMS1-mediated ubiquitination, thereby contributing to apoptotic susceptibility. In line, iASPP overexpression partially abolished the interaction of BRMS1 and CBP upon DNA damage. Reduced levels of iASPP mRNA and protein as well as CBP protein were observed in human melanoma compared with normal skin tissue and benign melanocytic nevi. In line with our findings, iASPP overexpression or knockdown of BRMS1 each augmented p300/CBP levels in melanoma cell lines, thereby enhancing apoptosis upon DNA damage. Taken together, destabilization of p300/CBP by downregulation of iASPP expression levels appears to represent a molecular mechanism that contributes to chemoresistance in melanoma cells.