Personalized computed tomography – Automated estimation of height and weight of a simulated digital twin using a 3D camera and artificial intelligence

Abstract

Purpose The aim of this study was to develop an algorithm for automated estimation of patient height and weight during computed tomography (CT) and to evaluate its accuracy in everyday clinical practice. Materials and methods Depth images of 200 patients were recorded with a 3D camera mounted above the patient table of a CT scanner. Reference values were obtained using a calibrated scale and a measuring tape to train a machine learning algorithm that fits a patient avatar into the recorded patient surface data. The resulting algorithm was prospectively used on 101 patients in clinical practice and the results were compared to the reference values and to estimates by the patient himself, the radiographer and the radiologist. The body mass index was calculated from the collected values for each patient using the WHO formula. A tolerance level of 5 kg was defined in order to evaluate the impact on weight-dependent contrast agent dosage in abdominal CT. Results Differences between values for height, weight and BMI were non-significant over all assessments (p > 0.83). The most accurate values for weight were obtained from the patient information (R² = 0.99) followed by the automated estimation via 3D camera (R² = 0.89). Estimates by medical staff were considerably less precise (radiologist: R² = 0.78, radiographer: R² = 0.77). A body-weight dependent dosage of contrast agent using the automated estimations matched the dosage using the reference measurements in 65 % of the cases. The dosage based on the medical staff estimates would have matched in 49 % of the cases. Conclusion Automated estimation of height and weight using a digital twin model from 3D camera acquisitions provide a high precision for protocol design in computer tomography. Key points: Citation Format

Patient digital twins: an introduction based on a scoping review

“…The first advantage of this definition is that it can exclude what is not a patient digital twin: generic models of cells, tissues, organs, or biological systems not linked to a patient but used to study disease progression or drug development 29,54,80 ; pure cyber-physical systems, that is, systems such as implantable cardioverter-defibrillators, which do not use a representation of the patient and therefore not a "viewable" digital replica of the patient; digital patient data created from patient databases for in silico trials; 47,99,103 often using generative adversarial networks, which we propose to call "synthetic patients" instead; data sets from another patient, similar to those of the index patient; 77 machine learning based classifiers, trained on a population to predict a diagnosis; 69 and patient models built from a single data source, such as demographic characteristics or imaging. 31,47,49 The second advantage of this definition is that it encompasses the two major trends in patient digital twins revealed by this study. On one hand, digital twins offering a high degree of fidelity, combining advanced anatomical and physiological models, based on mechanistic approaches or combining mechanistic and data-driven approaches.…”

The Digital Twin for Personalised Medicine: A Systematic Review

“…Ağırlık için en doğru tahminlerin sırasıyla hasta bilgileri, 3D kamerayla otomatik tahminden ve en az doğru tahminin tıbbi personele ait olduğu görülmüştür. Bu modelin iş yükünü, zamanı, maliyetleri ve hata kaynaklarını azaltabileceği belirtilmiştir (51).…”

“…Bunun yanı sıra hastalıklar gerçekleşmeden fark edilebilir, riskler azaltılabilir ve maliyetlerden tasarruf edilebilir (11). Yapılan çalışmalara bakıldığında birçok tıbbi alan ve hastalık için KT için Dİ'lerin kullanılabilir ve etkin olduğu görülmüştür (11,12,17,36,(42)(43)(44)(45)(46)(47)(48)(49)(50)(51).…”

Patient‐specific radiation risk‐based tube current modulation for diagnostic CT

“…To calculate the riskTCM curve, an estimation of the effective dose delivered per view to each organ has to be known beforehand, which requires a CT reconstruction of the patient. This could be achieved by using a deep learning-based reconstruction that utilizes the information from the available topogram(s) and a potential 3D camera image 15 of the patient surface.Such reconstructions based on one or two projections are described in the literature. [16][17][18][19][20] Since the focus of this study is to show the effective dose reduction potential of a riskminimizing TCM, CT reconstructions from a full scan are used instead.…”