Applied machine learning and artificial intelligence in rheumatology

Clinically relevant patient clusters identified by machine learning from the clinical development programme of secukinumab in psoriatic arthritis

Objectives

Identify distinct clusters of psoriatic arthritis (PsA) patients based on their baseline articular, entheseal and cutaneous disease manifestations and explore their clinical and therapeutic value.

Methods

Pooled baseline data in PsA patients (n=1894) treated with secukinumab across four phase 3 studies (FUTURE 2–5) were analysed to determine phenotypes based on clusters of clinical indicators. Finite mixture models methodology was applied to generate clinical clusters and mean longitudinal responses were compared between secukinumab doses (300 vs 150 mg) across identified clusters and clinical indicators through week 52 using machine learning (ML) techniques.

Results

Seven distinct patient clusters were identified. Cluster 1 (very-high (VH) – SWO/TEN (swollen/tender); n=187) was characterised by VH polyarticular burden for both tenderness and swelling of joints, while cluster 2 (H (high) – TEN; n=251) was marked by high polyarticular burden in tender joints and cluster 3 (H – Feet – Dactylitis; n=175) by high burden in joints of feet and dactylitis. For cluster 4 (L (Low) – Nails – Skin; n=209), cluster 5 (L – skin; n=283), cluster 6 (L – Nails; n=294) and cluster 7 (L; n=495) articular burden was low but nail and skin involvement was variable, with cluster 7 marked by mild disease activity across all domains. Greater improvements in the longitudinal responses for enthesitis in cluster 2, enthesitis and Psoriasis Area and Severity Index (PASI) in cluster 4 and PASI in cluster 6 were shown for secukinumab 300 mg compared with 150 mg.

Conclusions

PsA clusters identified by ML follow variable response trajectories indicating their potential to predict precise impact on patients’ outcomes.

Trial registration numbers

NCT01752634, NCT01989468, NCT02294227, NCT02404350

Accuracy and usability of a diagnostic decision support system in the diagnosis of three representative rheumatic diseases: a randomized controlled trial among medical students

BACKGROUND

An increasing number of diagnostic decision support systems (DDSS) exist to support patients and physicians in establishing the correct diagnosis as early as possible. However, little evidence exists that supports the effectiveness of these DDSS. The objectives were to compare the diagnostic accuracy of medical students, with and without the use of a DDSS, and the diagnostic accuracy of the DDSS system itself, regarding the typical rheumatic diseases and to analyze the user experience.

METHODS

A total of 102 medical students were openly recruited from a university hospital and randomized (unblinded) to a control group (CG) and an intervention group (IG) that used a DDSS (Ada - Your Health Guide) to create an ordered diagnostic hypotheses list for three rheumatic case vignettes. Diagnostic accuracy, measured as the presence of the correct diagnosis first or at all on the hypothesis list, was the main outcome measure and evaluated for CG, IG, and DDSS.

RESULTS

The correct diagnosis was ranked first (or was present at all) in CG, IG, and DDSS in 37% (40%), 47% (55%), and 29% (43%) for the first case; 87% (94%), 84% (100%), and 51% (98%) in the second case; and 35% (59%), 20% (51%), and 4% (51%) in the third case, respectively. No significant benefit of using the DDDS could be observed. In a substantial number of situations, the mean probabilities reported by the DDSS for incorrect diagnoses were actually higher than for correct diagnoses, and students accepted false DDSS diagnostic suggestions. DDSS symptom entry greatly varied and was often incomplete or false. No significant correlation between the number of symptoms extracted and diagnostic accuracy was seen. It took on average 7 min longer to solve a case using the DDSS. In IG, 61% of students compared to 90% in CG stated that they could imagine using the DDSS in their future clinical work life.

CONCLUSIONS

The diagnostic accuracy of medical students was superior to the DDSS, and its usage did not significantly improve students' diagnostic accuracy. DDSS usage was time-consuming and may be misleading due to prompting wrong diagnoses and probabilities.

TRIAL REGISTRATION

DRKS.de, DRKS00024433 . Retrospectively registered on February 5, 2021.

Personalized prediction of disease activity in patients with rheumatoid arthritis using an adaptive deep neural network

Background

Deep neural networks learn from former experiences on a large scale and can be used to predict future disease activity as potential clinical decision support. AdaptiveNet is a novel adaptive recurrent neural network optimized to deal with heterogeneous and missing clinical data.

Objective

We investigate AdaptiveNet for the prediction of individual disease activity in patients from a rheumatoid arthritis (RA) registry.

Methods

Demographic and disease characteristics from over 9500 patients and 65.000 visits from the Swiss Quality Management (SCQM) database were used to train and evaluate the network. Patient characteristics, clinical and patient reported outcomes, laboratory values and medication were used as input features. DAS28-BSR served as a target to predict active RA and future numeric individual disease activity by classification and regression.

Results

AdaptiveNet predicted active disease defined as DAS28-BSR >2.6 at the next visit with an overall accuracy of 75.6% (SD +- 0.7%) and a sensitivity and specificity of 84.2% (SD +- 1.6%) and 61.5% (SD +- 3.6%), respectively. Prediction performance was significantly higher in patients with a disease duration >3 years and positive rheumatoid factor. Regression allowed forecasting individual DAS28-BSR values with a mean squared error (MSE) of 0.9 (SD +- 0.05). This corresponds to a 8% deviation between estimated and real DAS28-BSR values. Compared to linear regression, random forest and support vector machines, AdaptiveNet showed an increased performance of over 7% in MSE. Medication played a minor role in the prediction of RA disease activity.

Conclusion

AdaptiveNet has a superior capacity to predict numeric RA disease activity compared to classical machine learning architectures. All investigated models had limitations in low specificity.

The Application and Development of Deep Learning in Radiotherapy: A Systematic Review

With the massive use of computers, the growth and explosion of data has greatly promoted the development of artificial intelligence (AI). The rise of deep learning (DL) algorithms, such as convolutional neural networks (CNN), has provided radiation oncologists with many promising tools that can simplify the complex radiotherapy process in the clinical work of radiation oncology, improve the accuracy and objectivity of diagnosis, and reduce the workload, thus enabling clinicians to spend more time on advanced decision-making tasks. As the development of DL gets closer to clinical practice, radiation oncologists will need to be more familiar with its principles to properly evaluate and use this powerful tool. In this paper, we explain the development and basic concepts of AI and discuss its application in radiation oncology based on different task categories of DL algorithms. This work clarifies the possibility of further development of DL in radiation oncology.

Machine Learning Solutions for Osteoporosis-A Review

Osteoporosis and its clinical consequence, bone fracture, is a multifactorial disease that has been the object of extensive research. Recent advances in machine learning (ML) have enabled the field of artificial intelligence (AI) to make impressive breakthroughs in complex data environments where human capacity to identify high-dimensional relationships is limited. The field of osteoporosis is one such domain, notwithstanding technical and clinical concerns regarding the application of ML methods. This qualitative review is intended to outline some of these concerns and to inform stakeholders interested in applying AI for improved management of osteoporosis. A systemic search in PubMed and Web of Science resulted in 89 studies for inclusion in the review. These covered one or more of four main areas in osteoporosis management: bone properties assessment (n = 13), osteoporosis classification (n = 34), fracture detection (n = 32), and risk prediction (n = 14). Reporting and methodological quality was determined by means of a 12-point checklist. In general, the studies were of moderate quality with a wide range (mode score 6, range 2 to 11). Major limitations were identified in a significant number of studies. Incomplete reporting, especially over model selection, inadequate splitting of data, and the low proportion of studies with external validation were among the most frequent problems. However, the use of images for opportunistic osteoporosis diagnosis or fracture detection emerged as a promising approach and one of the main contributions that ML could bring to the osteoporosis field. Efforts to develop ML-based models for identifying novel fracture risk factors and improving fracture prediction are additional promising lines of research. Some studies also offered insights into the potential for model-based decision-making. Finally, to avoid some of the common pitfalls, the use of standardized checklists in developing and sharing the results of ML models should be encouraged. © 2021 American Society for Bone and Mineral Research (ASBMR).

The Role of Artificial Intelligence in the Diagnosis and Prognosis of Renal Cell Tumors

The increasing availability of molecular data provided by next-generation sequencing (NGS) techniques is allowing improvement in the possibilities of diagnosis and prognosis in renal cancer. Reliable and accurate predictors based on selected gene panels are urgently needed for better stratification of renal cell carcinoma (RCC) patients in order to define a personalized treatment plan. Artificial intelligence (AI) algorithms are currently in development for this purpose. Here, we reviewed studies that developed predictors based on AI algorithms for diagnosis and prognosis in renal cancer and we compared them with non-AI-based predictors. Comparing study results, it emerges that the AI prediction performance is good and slightly better than non-AI-based ones. However, there have been only minor improvements in AI predictors in terms of accuracy and the area under the receiver operating curve (AUC) over the last decade and the number of genes used had little influence on these indices. Furthermore, we highlight that different studies having the same goal obtain similar performance despite the fact they use different discriminating genes. This is surprising because genes related to the diagnosis or prognosis are expected to be tumor-specific and independent of selection methods and algorithms. The performance of these predictors will be better with the improvement in the learning methods, as the number of cases increases and by using different types of input data (e.g., non-coding RNAs, proteomic and metabolic). This will allow for more precise identification, classification and staging of cancerous lesions which will be less affected by interpathologist variability.

Complex Machine-Learning Algorithms and Multivariable Logistic Regression on Par in the Prediction of Insufficient Clinical Response to Methotrexate in Rheumatoid Arthritis

The goals of this study were to examine whether machine-learning algorithms outperform multivariable logistic regression in the prediction of insufficient response to methotrexate (MTX); secondly, to examine which features are essential for correct prediction; and finally, to investigate whether the best performing model specifically identifies insufficient responders to MTX (combination) therapy. The prediction of insufficient response (3-month Disease Activity Score 28-Erythrocyte-sedimentation rate (DAS28-ESR) > 3.2) was assessed using logistic regression, least absolute shrinkage and selection operator (LASSO), random forest, and extreme gradient boosting (XGBoost). The baseline features of 355 rheumatoid arthritis (RA) patients from the “treatment in the Rotterdam Early Arthritis CoHort” (tREACH) and the U-Act-Early trial were combined for analyses. The model performances were compared using area under the curve (AUC) of receiver operating characteristic (ROC) curves, 95% confidence intervals (95% CI), and sensitivity and specificity. Finally, the best performing model following feature selection was tested on 101 RA patients starting tocilizumab (TCZ)-monotherapy. Logistic regression (AUC = 0.77 95% CI: 0.68–0.86) performed as well as LASSO (AUC = 0.76, 95% CI: 0.67–0.85), random forest (AUC = 0.71, 95% CI: 0.61 = 0.81), and XGBoost (AUC = 0.70, 95% CI: 0.61–0.81), yet logistic regression reached the highest sensitivity (81%). The most important features were baseline DAS28 (components). For all algorithms, models with six features performed similarly to those with 16. When applied to the TCZ-monotherapy group, logistic regression’s sensitivity significantly dropped from 83% to 69% (p = 0.03). In the current dataset, logistic regression performed equally well compared to machine-learning algorithms in the prediction of insufficient response to MTX. Models could be reduced to six features, which are more conducive for clinical implementation. Interestingly, the prediction model was specific to MTX (combination) therapy response.

Assessing the risk of dengue severity using demographic information and laboratory test results with machine learning

BACKGROUND

Dengue virus causes a wide spectrum of disease, which ranges from subclinical disease to severe dengue shock syndrome. However, estimating the risk of severe outcomes using clinical presentation or laboratory test results for rapid patient triage remains a challenge. Here, we aimed to develop prognostic models for severe dengue using machine learning, according to demographic information and clinical laboratory data of patients with dengue.

METHODOLOGY/PRINCIPAL FINDINGS

Out of 1,581 patients in the National Cheng Kung University Hospital with suspected dengue infections and subjected to NS1 antigen, IgM and IgG, and qRT-PCR tests, 798 patients including 138 severe cases were enrolled in the study. The primary target outcome was severe dengue. Machine learning models were trained and tested using the patient dataset that included demographic information and qualitative laboratory test results collected on day 1 when they sought medical advice. To develop prognostic models, we applied various machine learning methods, including logistic regression, random forest, gradient boosting machine, support vector classifier, and artificial neural network, and compared the performance of the methods. The artificial neural network showed the highest average discrimination area under the receiver operating characteristic curve (0.8324 ± 0.0268) and balance accuracy (0.7523 ± 0.0273). According to the model explainer that analyzed the contributions/co-contributions of the different factors, patient age and dengue NS1 antigenemia were the two most important risk factors associated with severe dengue. Additionally, co-existence of anti-dengue IgM and IgG in patients with dengue increased the probability of severe dengue.

CONCLUSIONS/SIGNIFICANCE

We developed prognostic models for the prediction of dengue severity in patients, using machine learning. The discriminative ability of the artificial neural network exhibited good performance for severe dengue prognosis. This model could help clinicians obtain a rapid prognosis during dengue outbreaks. However, the model requires further validation using external cohorts in future studies.

[Perspectives for rheumatological health services research at the German Rheumatism Research Center]

In diesem Übersichtsartikel werden aktuelle Projekte und Perspektiven der rheumatologischen Versorgungsforschung am Programmbereich Epidemiologie des DRFZ (Deutsches Rheuma-Forschungszentrum) zusammengefasst. Versorgungsforschung wird mithilfe verschiedener Datenquellen betrieben. Neben den klassischen rheumatologischen Krankheitsregistern werden zunehmend auch Krankenkassendaten und bevölkerungsbezogene Kohorten für Analysen verwendet. Von der Datenerfassung über das Monitoring bis zu Analysealgorithmen verändern digitale Anwendungen die Versorgungsforschung der nächsten Jahre. Kollaborative Analysen mit nationalen und internationalen Kooperationspartnern unter Einbindung von Biomarkern komplettieren die Forschung am Programmbereich Epidemiologie. Die Digitalisierung der Forschungsprojekte ist ein zentraler Baustein, der die Versorgungsforschung im kommenden Jahrzehnt weiter verändern wird.

Artificial intelligence and the hunt for immunological disorders

PURPOSE OF REVIEW

Artificial intelligence has pervasively transformed many industries and is beginning to shape medical practice. New use cases are being identified in subspecialty domains of medicine and, in particular, application of artificial intelligence has found its way to the practice of allergy-immunology. Here, we summarize recent developments, emerging applications and obstacles to realizing full potential.

RECENT FINDINGS

Artificial/augmented intelligence and machine learning are being used to reduce dimensional complexity, understand cellular interactions and advance vaccine work in the basic sciences. In genomics, bioinformatic methods are critical for variant calling and classification. For clinical work, artificial intelligence is enabling disease detection, risk profiling and decision support. These approaches are just beginning to have impact upon the field of clinical immunology and much opportunity exists for further advancement.

SUMMARY

This review highlights use of computational methods for analysis of large datasets across the spectrum of research and clinical care for patients with immunological disorders. Here, we discuss how big data methods are presently being used across the field clinical immunology.

Digital crowdsourcing: unleashing its power in rheumatology

The COVID-19 pandemic forces the whole rheumatic and musculoskeletal diseases community to reassemble established treatment and research standards. Digital crowdsourcing is a key tool in this pandemic to create and distil desperately needed clinical evidence and exchange of knowledge for patients and physicians alike. This viewpoint explains the concept of digital crowdsourcing and discusses examples and opportunities in rheumatology. First experiences of digital crowdsourcing in rheumatology show transparent, accessible, accelerated research results empowering patients and rheumatologists.