Noncontact Physiological Measurement Using a Camera: A Technical Review and Future Directions.

Using a camera as an optical sensor to monitor physiological parameters has garnered considerable research interest in biomedical engineering in recent decades. Researchers have explored the use of a camera for monitoring a variety of physiological waveforms, together with the vital signs carried by these waveforms. Most of the obtained waveforms are related to the human respiratory and cardiovascular systems, and in addition of being indicative of overall health, they can also detect early signs of certain diseases. While using a camera for noncontact physiological signal monitoring offers the advantages of low cost and operational ease, it also has the disadvantages such as vulnerability to motion and lack of burden-free calibration solutions in some use cases. This study presents an overview of the existing camera-based methods that have been reported in recent years. It introduces the physiological principles behind these methods, signal acquisition approaches, various types of acquired signals, data processing algorithms, and application scenarios of these methods. It also discusses the technological gaps between the camera-based methods and traditional medical techniques, which are mostly contact-based. Furthermore, we present the manner in which noncontact physiological signal monitoring use has been extended, particularly over the recent years, to more day-to-day aspects of individuals' lives, so as to go beyond the more conventional use case scenarios. We also report on the development of novel approaches that facilitate easier measurement of less often monitored and recorded physiological signals. These have the potential of ushering a host of new medical and lifestyle applications. We hope this study can provide useful information to the researchers in the noncontact physiological signal measurement community.

The value of the computer-aided diagnosis system for thyroid lesions based on computed tomography images.

BACKGROUND: Thyroid nodules are commonly found at palpation amounting to 4-7% of the asymptomatic population and 50% of the cases are found at autopsy. Only a small proportion of thyroid nodules are malignant. The major challenge is the differential diagnosis of benign or malignant thyroid nodules, so we aim to develop the computer-assisted diagnostic method based on computed tomography (CT) images for thyroid lesions. METHODS: In this study, we retrospectively collected 52 benign and 46 malignant thyroid nodules from 90 patients in CT examinations, together with the pathologist findings and radiology diagnosis. The first-order statistic and gray-level co-occurrence matrix features were extracted from thyroid computed tomography images. These texture features were used to assess the malignancy risk of the thyroid nodules. Several classification algorithms, including support vector machine, linear discriminant analysis, random forest, and bootstrap aggregating, were applied in the prediction. Leave-one-out cross-validation was used to evaluate the performance of thyroid cancer recognition. RESULTS: In thyroid cancer identification based on a computed tomography image, we found the system using 17 texture features and support vector machine performed well. The accuracy, area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value, were 0.8673, 0.9105, 0.9130, 0.8269, 0.8235 and 0.9146, respectively. CONCLUSIONS: The proposed computer-aided diagnosis system provides a good assessment of the malignancy-risk of the thyroid nodules, which may help radiologists to improve the accuracy and efficiency of thyroid diagnosis.

Thyroid nodule recognition in computed tomography using first order statistics.

BACKGROUND: Computed tomography (CT) is one of the popular tools for early detection of thyroid nodule. The pixel intensity of thyroid in CT image is very important information to distinguish nodule from normal thyroid tissue. The pixel intensity in normal thyroid tissues is homogeneous and smooth. In the benign or malignant nodules, the pixel intensity is heterogeneous. Several studies have shown that the first order features in ultrasound image can be used as imaging biomarkers in nodule recognition. METHODS: In this paper, we investigate the feasibility of utilizing the first order texture features to identify nodule from normal thyroid tissue in CT image. A total of 284 thyroid CT images from 113 patients were collected in this study. We used 150 healthy controlled thyroid CT images from 55 patients and 134 nodule images (50 malignant and 84 benign nodules) from 58 patients who have undergone thyroid surgery. The final diagnosis was confirmed by histopathological examinations. In the presented method, first, regions of interest (ROIs) from axial non-enhancement CT images were delineated manually by a radiologist. Second, average, median, and wiener filter were applied to reduce photon noise before feature extraction. The first-order texture features, including entropy, uniformity, average intensity, standard deviation, kurtosis and skewness were calculated from each ROI. Third, support vector machine analysis was applied for classification. Several statistical values were calculated to evaluate the performance of the presented method, which includes accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and area of under receiver operating characteristic curve (AUC). RESULTS: The entropy, uniformity, mean intensity, standard deviation, skewness (P < 0.05), except kurtosis (P = 0.104) of thyroid tissue with nodules have a significant difference from those of normal thyroid tissue. The optimal classification was obtained from the presented method. The accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) are 0.880, 0.821, 0.933, 0.917, 0.854, and 0.953 respectively. CONCLUSION: First order texture features can be used as imaging biomarkers, and the presented system can be used to assist radiologists to recognize the nodules in CT image.

Noncontact spirometry with a webcam.

We present an imaging-based method for noncontact spirometry. The method tracks the subtle respiratory-induced shoulder movement of a subject, builds a calibration curve, and determines the flow-volume spirometry curve and vital respiratory parameters, including forced expiratory volume in the first second, forced vital capacity, and peak expiratory flow rate. We validate the accuracy of the method by comparing the data with those simultaneously recorded with a gold standard reference method and examine the reliability of the noncontact spirometry with a pilot study including 16 subjects. This work demonstrates that the noncontact method can provide accurate and reliable spirometry tests with a webcam. Compared to the traditional spirometers, the present noncontact spirometry does not require using a spirometer, breathing into a mouthpiece, or wearing a nose clip, thus making spirometry test more easily accessible for the growing population of asthma and chronic obstructive pulmonary diseases.

Simultaneous Monitoring of Ballistocardiogram and Photoplethysmogram Using a Camera.

We present a noncontact method to measure ballistocardiogram (BCG) and photoplethysmogram (PPG) simultaneously using a single camera. The method tracks the motion of facial features to determine displacement BCG, and extracts the corresponding velocity and acceleration BCGs by taking first and second temporal derivatives from the displacement BCG, respectively. The measured BCG waveforms are consistent with those reported in the literature and also with those recorded with an accelerometer-based reference method. The method also tracks PPG based on the reflected light from the same facial region, which makes it possible to track both BCG and PPG with the same optics. We verify the robustness and reproducibility of the noncontact method with a small pilot study with 23 subjects. The presented method is the first demonstration of simultaneous BCG and PPG monitoring without wearing any extra equipment or marker by the subject.

Motion robust remote photoplethysmography in CIELab color space.

Remote photoplethysmography (rPPG) is attractive for tracking a subject's physiological parameters without wearing a device. However, rPPG is known to be prone to body movement-induced artifacts, making it unreliable in realistic situations. Here we report a method to minimize the movement-induced artifacts. The method selects an optimal region of interest (ROI) automatically, prunes frames in which the ROI is not clearly captured (e.g., subject moves out of the view), and analyzes rPPG using an algorithm in CIELab color space, rather than the widely used RGB color space. We show that body movement primarily affects image intensity, rather than chromaticity, and separating chromaticity from intensity in CIELab color space thus helps achieve effective reduction of the movement-induced artifacts. We validate the method by performing a pilot study including 17 people with diverse skin tones.

Noncontact Monitoring of Blood Oxygen Saturation Using Camera and Dual-Wavelength Imaging System.

We present a noncontact method to monitor blood oxygen saturation (SpO2). The method uses a CMOS camera with a trigger control to allow recording of photoplethysmography (PPG) signals alternatively at two particular wavelengths, and determines the SpO2 from the measured ratios of the pulsatile to the nonpulsatile components of the PPG signals at these wavelengths. The signal-to-noise ratio (SNR) of the SpO2 value depends on the choice of the wavelengths. We found that the combination of orange (λ = 611 nm) and near infrared (λ = 880 nm) provides the best SNR for the noncontact video-based detection method. This combination is different from that used in traditional contact-based SpO 2 measurement since the PPG signal strengths and camera quantum efficiencies at these wavelengths are more amenable to SpO2 measurement using a noncontact method. We also conducted a small pilot study to validate the noncontact method over an SpO2 range of 83%-98%. This study results are consistent with those measured using a reference contact SpO2 device ( r = 0.936, ). The presented method is particularly suitable for tracking one's health and wellness at home under free-living conditions, and for those who cannot use traditional contact-based PPG devices.

Noncontact monitoring breathing pattern, exhalation flow rate and pulse transit time.

We present optical imaging-based methods to measure vital physiological signals, including breathing frequency (BF), exhalation flow rate, heart rate (HR), and pulse transit time (PTT). The breathing pattern tracking was based on the detection of body movement associated with breathing using a differential signal processing approach. A motion-tracking algorithm was implemented to correct random body movements that were unrelated to breathing. The heartbeat pattern was obtained from the color change in selected region of interest (ROI) near the subject's mouth, and the PTT was determined by analyzing pulse patterns at different body parts of the subject. The measured BF, exhaled volume flow rate and HR are consistent with those measured simultaneously with reference technologies (r = 0.98, for HR; r = 0.93, for breathing rate), and the measured PTT difference (30-40 ms between mouth and palm) is comparable to the results obtained with other techniques in the literature. The imaging-based methods are suitable for tracking vital physiological parameters under free-living condition and this is the first demonstration of using noncontact method to obtain PTT difference and exhalation flow rate.