Smartphone Lens Research Articles

Methods of Assessing Nailfold Capillaroscopy Compared to Video Capillaroscopy in Patients with Systemic Sclerosis—A Critical Review of the Literature

Introduction: Nailfolds of patients with systemic sclerosis (SSc) provide an opportunity to directly visualize microvascular remodeling in SSc. Nailfold video capillaroscopy (NVC) remains the gold standard for assessing nailfold capillaroscopy (NFC). However, access to NVC is limited by expense and expertise. This review aims to synthesize current research on other NFC devices compared to NVC. Methods: The literature search included the primary research of adult patients with SSc as defined by the 2013 ACR/EULAR criteria. Methods of assessing NFC included stereomicroscopy/wide-field microscopy, ophthalmoscopy, dermatoscopy, smartphone devices, and digital USB microscopy. Primary outcomes included both qualitative (normal vs. abnormal nailfolds, overall pattern recognition, presence/absence of giant capillaries, hemorrhages, and abnormal morphology) and quantitative (capillary density and dimension) measures. Results: The search yielded 471 studies, of which 9 were included. Five studies compared NVC to dermatoscopy, two compared it to widefield/stereomicroscopy, one to smartphone attachments, and one to USB microscopy. In dermatoscopy studies, NVC had a higher percentage of images that were interpretable (63–77% vs. 100%), classifiable (70% vs. 84%), or gradable (70% vs. 79.3%) across three studies. Dermatoscopy had a lower sensitivity (60.2% vs. 81.6%) and higher specificity (92.5% vs. 84.6%) compared to NVC. One stereomicroscopy study found a significant difference between methods in capillary density in limited cutaneous SSc, while another found correlations in all parameters between stereomicroscopy and NVC. One smartphone lens had good agreement with NVC on abnormal capillary morphology and density. USB microscopy was able to differentiate between SSc and healthy controls using mean capillary width but not by capillary density. Discussion: A dermatoscope may serve as a more portable and affordable screening tool to identify a normal “scleroderma pattern”, and images that need further corroboration by NVC. NFC parameters reported are heterogenous and the standardization of these parameters is important, especially in non-gold-standard devices.

A smartphone lens attachment improves the quality of referrals to eye casualty.

Background/objectivesIn recent years, eye casualty clinics have seen significant increases in patient numbers with reduced capacity. COVID-19 has exacerbated this issue and demonstrated the potential of telemedicine as a solution. Our study evaluated the potential benefit of a smartphone-based lens attachment to improve the referral pathway for anterior segment related complaints in eye casualty.Subjects/methodsFifty-four consecutive patients with anterior segment complaints were recruited. A questionnaire was completed with each patient to simulate the history from the point of referral. White light and cobalt blue photos were captured using a smartphone lens. The clinician reviewing the patient was asked to document the actual diagnosis and the appropriate time-frame within which they felt the patient could safely have been seen within; both with and without the option of management advice at the time of triage. The subsequent images and questionnaires were reviewed by a single consultant Ophthalmologist who was independent to the data collection process. The assessor was asked to make a diagnosis and management plan based upon the questionnaire (‘History’), and the questionnaire with the photo (‘History with Image’), as well as rate their clinical confidence on a 1–5 scale.ResultsDiagnostic accuracy was significantly higher in “History with Image” (98.2%), when compared to “History” only (48.2%). “History with Image” prevented significantly more appointments when compared to “History” alone, at similar levels to retrospective clinic review. Preventable appointments were increased if clinical advice was possible. Timeframe of appointments between ‘History with Image’ and ‘Clinic’ appointments was similar. Clinical’Confidence was significantly higher at 4.5 with ‘History with Image’ when compared to 2.37 with ‘History Only’.ConclusionA low-cost smartphone lens attachment, alongside a standardised clinical questionnaire, can improve the referral pathway to the hospital eye service by reducing unnecessary appointments, while improving clinical confidence and diagnostic accuracy during triage. Further work to evaluate referral pathways, including the development of systems that allow for secure image transmission are needed to understand the feasibility for the widespread adoption of this technology.

Automated Digitization of the Cardiotocography Signals from Real Scene Image of Binary Clinic Report

ObjectiveCardiotocography (CTG) is a popular clinical tool to assess foetal health status. Automatic CTG analysis can remove the judgement differences of inter- and intra- doctors. However, absence of CTG database has hindered the development of automatic CTG analysis based on deep learning. Therefore, digitization of CTG signal from clinic report is an important way to enrich CTG database, and in turn promote the automatic CTG analysis. MethodThe proposed digitization method extracts digital signal from the commonly used binary CTG paper reports. An adaptive region positioning algorithm based on statistical calculation is used to locate signal regions. Then, by deducing the dynamic minimal weight sum in theory, methods based on statistical calculation and on the weight sum are designed to remove grid lines. Next, according to different breakpoint types and signal line trends, different signal reconstruction methods are formulated. This realize extracting the signal line from binary background grid lines. Finally, a calibration method based on segmented sampling is designed to reduce calibration error due to smartphone lens distortion. ResultsThe experimental results show that the correlation coefficients of FHR and UC of proposed method both reached 0.98. After three expert gynaecologists's evaluations, there are no clinically relevant differences were identified between the extracted signals and the reference ones.

Smartphone-based corneal topography with null-screens.

To measure the shape of the fast corneal surface of the human eye, we propose the design and characterization of a compact corneal topographer using the capabilities of a smartphone. The performance evaluation of the compact corneal topographer includes the calculation and compensation of the distortion introduced by the smartphone lens used to acquire the images and the evaluation of a reference surface. To demonstrate the feasibility of our proposal, we performed surface topography measurements on some human corneas and compared the results with those obtained by a commercial corneal topographer. We showed that the results obtained with our algorithms were consistent with other methods that analyze the corneal surface.

A novel diversity method for smartphone camera-based heart rhythm signals in the presence of motion and noise artifacts.

The advent of smartphones has advanced the use of embedded sensors to acquire various physiological information. For example, smartphone camera sensors and accelerometers can provide heart rhythm signals to the subjects, while microphones can give respiratory signals. However, the acquired smartphone-based physiological signals are more vulnerable to motion and noise artifacts (MNAs) compared to using medical devices, since subjects need to hold the smartphone with proper contact to the smartphone camera and lens stably and tightly for a duration of time without any movement in the hand or finger. This results in more MNA than traditional methods, such as placing a finger inside a tightly enclosed pulse oximeter to get PPG signals, which provides stable contact between the sensor and the subject’s finger. Moreover, a smartphone lens does not block ambient light in an effective way, while pulse oximeters are designed to block the ambient light effectively. In this paper, we propose a novel diversity method for smartphone signals that reduces the effect of MNAs during heart rhythm signal detection by 1) acquiring two heterogeneous signals from a color intensity signal and a fingertip movement signal, and 2) selecting the less MNA-corrupted signal of the two signals. The proposed method has advantages in that 1) diversity gain can be obtained from the two heterogeneous signals when one signal is clean while the other signal is corrupted, and 2) acquisition of the two heterogeneous signals does not double the acquisition procedure but maintains a single acquisition procedure, since two heterogeneous signals can be obtained from a single smartphone camera recording. In our diversity method, we propose to choose the better signal based on the signal quality indices (SQIs), i.e., standard deviation of instantaneous heart rate (STD–HR), root mean square of the successive differences of peak-to-peak time intervals (RMSSD–T), and standard deviation of peak values (STD–PV). As a performance metric evaluating the proposed diversity method, the ratio of usable period is considered. Experimental results show that our diversity method increases the usable period 19.53% and 6.25% compared to the color intensity or the fingertip movement signals only, respectively.

Shoreline change mapping using crowd-sourced smartphone images

Shoreline change information is critical for effective management of the coastal zone. This study presents a low-cost method for mapping shoreline change that harnesses smartphone images collected by the community and uploaded to social media platforms. A smartphone camera cradle installed overlooking a coastal region is used to constrain the crowd-sourced camera extrinsic parameters and accompanying signage instructs participants on how to share their image to social media using a site-specific hashtag identifier. Surveyed ground control points solve for the focal length of the smartphone lens and more-accurately resolve the camera extrinsic parameters. Shoreline position is subsequently mapped on georectified images using an edge detection technique based on the red and blue colour channels. A validation of the method was conducted at two sandy beaches in SE Australia and resulted in strong community participation (400 images submitted over 7 months by 198 individual contributors). Concurrent shoreline surveys using RTK-GNSS indicated that shoreline accuracy using this crowd-sourced approach is comparable to that of established coastal imaging systems, with cross-shore shoreline accuracy best for these two elevated validation sites (camera elevation = 17.3 m–27.1 m above MSL) in the camera nearfield (RMSD ≈ 1.4 m) and RMSD ranging between 2.6 and 3.9 m over coastal stretches spanning up to 1 km. Minimal differences in shoreline accuracy were observed between low resolution images characteristic of those uploaded to social media and higher resolution images sourced from the smartphone. The successful application of this low-cost approach, combined with the proliferation of smartphones and social media usage, open up new possibilities for crowd-sourced shoreline change mapping at suitable coastal locations worldwide.

Open-source, machine and deep learning-based automated algorithm for gestational age estimation through smartphone lens imaging.

Gestational age estimation at time of birth is critical for determining the degree of prematurity of the infant and for administering appropriate postnatal treatment. We present a fully automated algorithm for estimating gestational age of premature infants through smartphone lens imaging of the anterior lens capsule vasculature (ALCV). Our algorithm uses a fully convolutional network and blind image quality analyzers to segment usable anterior capsule regions. Then, it extracts ALCV features using a residual neural network architecture and trains on these features using a support vector machine-based classifier. The classification algorithm is validated using leave-one-out cross-validation on videos captured from 124 neonates. The algorithm is expected to be an influential tool for remote and point-of-care gestational age estimation of premature neonates in low-income countries. To this end, we have made the software open source.

Using machine-learning to optimize phase contrast in a low-cost cellphone microscope.

Cellphones equipped with high-quality cameras and powerful CPUs as well as GPUs are widespread. This opens new prospects to use such existing computational and imaging resources to perform medical diagnosis in developing countries at a very low cost. Many relevant samples, like biological cells or waterborn parasites, are almost fully transparent. As they do not exhibit absorption, but alter the light’s phase only, they are almost invisible in brightfield microscopy. Expensive equipment and procedures for microscopic contrasting or sample staining often are not available. Dedicated illumination approaches, tailored to the sample under investigation help to boost the contrast. This is achieved by a programmable illumination source, which also allows to measure the phase gradient using the differential phase contrast (DPC) [1, 2] or even the quantitative phase using the derived qDPC approach [3]. By applying machine-learning techniques, such as a convolutional neural network (CNN), it is possible to learn a relationship between samples to be examined and its optimal light source shapes, in order to increase e.g. phase contrast, from a given dataset to enable real-time applications. For the experimental setup, we developed a 3D-printed smartphone microscope for less than 100 $ using off-the-shelf components only such as a low-cost video projector. The fully automated system assures true Koehler illumination with an LCD as the condenser aperture and a reversed smartphone lens as the microscope objective. We show that the effect of a varied light source shape, using the pre-trained CNN, does not only improve the phase contrast, but also the impression of an improvement in optical resolution without adding any special optics, as demonstrated by measurements.

Comparison Study of Funduscopic Exam of Pediatric Patients Using the D-EYE Method and Conventional Indirect Ophthalmoscopic Methods

Purpose: The D-EYE device, a new fundoscopic smartphone lens, has demonstrated its utility in a clinical setting to detect and document ocular pathology, but has not been tested in the pediatric population. A prospective study was performed to explore the application of D-EYE in pediatric fundus examinations. Methods: Patients ages 3 - 18 years old underwent dilated fundus examinations by masked examiners using the video function of the D-EYE, while indirect ophthalmoscopy was performed by apediatric ophthalmologist. The examiners independently analyzed the D-EYE videos for the presence or absence of abnormalities, cup-to-disc (c/d) ratios and optic nerve size and color. The D-EYE video findings were compared to indirect ophthalmoscopy findings. Results: The study included 172 eyes from 87 patients. In comparing D-EYE to indirect ophthalmoscopy for detecting fundus abnormalities, the sensitivity was 0.72, specificity was 0.97, positive predictive value (PPV) was 0.77, negative predictive value (NPV) was 0.97, positive likelihood ratio (LR) was 27.8, and negative LR was 0.29. The agreement rate between the D-EYE video graders for the c/d ratio within a value of 0.1 was 97.0%. Multiple, distinct abnormalities were discovered using the D-EYE device, including nystagmus, optic nerve hypoplasia, optic disc edema, peripapillary atrophy, disc pallor, and optic disc drusen. Conclusion: Fundoscopic imaging using the D-EYE smartphone lens reliably detects the presence of fundus abnormalities and has good reliability in assessing c/d ratios. The video capability is useful in patients with nystagmus or those who are poorly compliant with the examination and allowed for effective teaching by the pediatric ophthalmologist.