Search

Oscillometric monitors are the most common automated blood pressure (BP) measurement devices used in non-specialist settings. However, their accuracy and reliability vary under different settings and for different age groups and health conditions. A main limitation of the existing oscillometric monitors is their underlying analysis algorithms that are unable to fully capture the BP information encoded in the pattern of the recorded oscillometric pulses. In this paper, we propose a new 2D oscillometric data representation that enables a full characterization of arterial system and empowers the application of deep learning to extract the most informative features correlated with BP. A hybrid convolutional-recurrent neural network was developed to capture the oscillometric pulses morphological information as well as their temporal evolution over the cuff deflation period from the 2D structure, and estimate BP. The performance of the proposed method was verified on three oscillometric databases collected from the wrist and upper arms of 245 individuals. It was found that it achieves a mean error and a standard deviation of error of as low as 0.08 mmHg and 2.4 mmHg in the estimation of systolic BP, and 0.04 mmHg and 2.2 mmHg in the estimation of diastolic BP, respectively. Our proposed method outperformed the state-of-the-art techniques and satisfied the current international standards for BP monitors by a wide margin. The proposed method shows promise toward robust and objective BP estimation in a variety of patients and monitoring situations.

Segmentation of lung tissue in computed tomography (CT) images is a precursor to most pulmonary image analysis applications. Semantic segmentation methods using deep learning have exhibited top-tier performance in recent years, however designing accurate and robust segmentation models for lung tissue is challenging due to the variations in shape, size, and orientation. Additionally, medical image artifacts and noise can affect lung tissue segmentation and degrade the accuracy of downstream analysis. The practicality of current deep learning methods for lung tissue segmentation is limited as they require significant computational resources and may not be easily deployable in clinical settings. This paper presents a fully automatic method that identifies the lungs in three-dimensional (3D) pulmonary CT images using deep networks and transfer learning. We introduce (1) a novel 2.5-dimensional image representation from consecutive CT slices that succinctly represents volumetric information and (2) a U-Net architecture equipped with pre-trained InceptionV3 blocks to segment 3D CT scans while maintaining the number of learnable parameters as low as possible. Our method was quantitatively assessed using one public dataset, LUNA16, for training and testing and two public datasets, namely, VESSEL12 and CRPF, only for testing. Due to the low number of learnable parameters, our method achieved high generalizability to the unseen VESSEL12 and CRPF datasets while obtaining superior performance over Luna16 compared to existing methods (Dice coefficients of 99.7, 99.1, and 98.8 over LUNA16, VESSEL12, and CRPF datasets, respectively). We made our method publicly accessible via a graphical user interface at medvispy.ee.kntu.ac.ir.

Oscillometry is the most common measurement method used in automatic blood pressure (BP) monitors. However, most of the oscillometric algorithms are without physiological and theoretical foundation, and rely on empirically derived coefficients for systolic and diastolic pressure evaluation which affects the reliability of the technique. In this thesis, the oscillometric BP estimation problem is addressed using a comprehensive modeling approach, based on which coefficient-free estimation of BP becomes possible. A feature-based neural network approach is developed to find an implicit relationship between BP and the oscillometric waveform (OMW). The modeling approach is then extended by developing a mathematical model for the OMW as a function of the arterial blood pressure, cuff pressure, and cuff-arm-artery system parameters. Based on the developed model, the explicit relationship between the OMW and the systolic and diastolic pressures is found and a new coefficient-free oscillometric BP estimation method using the trust region reflective algorithm is proposed. In order to improve the reliability of BP estimates, the electrocardiogram signal is recorded simultaneously with the OMW, as another independent source of information. The electrocardiogram signal is used to identify the true oscillometric pulses and calculate the pulse transit time (PTT). By combining our developed model of oscillomtery with an existing model of the pulse wave velocity, a new mathematical model is derived for the PTT during the cuff deflation. The derived model is incorporated to study the PTT-cuff pressure dependence, based on which a new coefficient-free BP estimation method is proposed. In order to obtain accurate and robust estimates of BP, the proposed model-based BP estimation method sare fused by computing the weighted arithmetic mean of their estimates. With fusion of the proposed methods, it is observed that the mean absolute error (MAE) in estimation of systolic and diastolic pressures is 4.40 and 3.00 mmHg, respectively, relative to the Food and Drug Administration-approved Omron monitor. In addition, the proposed feature-based neural network was compared with auscultatory measurements by trained observers giving MAE of 6.28 and 5.73 mmHg in estimation of systolic and diastolic pressures, respectively. The proposed models thus show promise toward developing robust BP estimation methods.