Macular edema (ME) and central serous retinopathy (CSR) are two macular diseases that affect the central vision of a person if they are left untreated. Optical coherence tomography (OCT) imaging is the latest eye examination technique that shows a cross-sectional region of the retinal layers and that can be used to detect many retinal disorders in an early stage. Many researchers have done clinical studies on ME and CSR and reported significant findings in macular OCT scans. However, this paper proposes an automated method for the classification of ME and CSR from OCT images using a support vector machine (SVM) classifier. Five distinct features (three based on the thickness profiles of the sub-retinal layers and two based on cyst fluids within the sub-retinal layers) are extracted from 30 labeled images (10 ME, 10 CSR, and 10 healthy), and SVM is trained on these. We applied our proposed algorithm on 90 time-domain OCT (TD-OCT) images (30 ME, 30 CSR, 30 healthy) of 73 patients. Our algorithm correctly classified 88 out of 90 subjects with accuracy, sensitivity, and specificity of 97.77%, 100%, and 93.33%, respectively.

Lung cancer is currently one of the most common causes of death worldwide, with low rates of survival after diagnosis being reported in developed and underdeveloped countries.¹ According to recent statistics, the 5-year survival rate is only 16%² and it has been estimated that by the year 2020, 12 million cancer-related fatalities will occur annually, of which lung cancer will have the largest share.³ However, survival rates can be improved,¹ if nodules are detected early enough. Lung nodules are abnormal growths of tissue that could represent lung cancer. They are typically round/spherical in shape with diameters of up to 30 mm.⁴ Nodules are categorized as well circumscribed, juxtavascular, juxtapleural, and pleural tail. Well-circumscribed nodules are independent and have no extensions into the surrounding anatomical structures, whereas juxtavascular nodules exhibit strong adherence to proximal vessels, and juxtapleural nodules are attached to neighboring pleural surfaces. Pleural-tail nodules have tails that are adherent to the nodule but not to pleural walls. In addition, pulmonary nodules are categorized as solid and subsolid nodules (SSNs), irrespective of their positions. SSNs are further classified as part-solid nodules and pure ground-glass nodules. Solid nodules are the most common type of nodules, and these repress the underlying functional lung tissues. SSNs are pulmonary nodules with partial ground-glass opacity (GGO). These nodules exhibit opacifications with higher density than the surrounding tissues and do not obscure underlying bronchovascular structures.⁵ Sample images of different nodules are shown in Fig. 1.

Self-mixing or optical feedback interferometry has been widely used for displacement and velocity measurement applications. For metric information retrieval with <; λ/2 precision, various phase unwrapping methods have been proposed. However, these are computationally heavy and require large number of hardware resources, thereby hindering the development of real-time, embedded solutions for large bandwidth applications. In this regard, a simple and efficient feedback phase retrieval algorithm, called consecutive samples-based unwrapping (CSU) is presented. Detailed analysis of its error performance has been conducted as a function of key optical feedback parameters. A theoretical study has also been conducted to explain as to why such good error performance is obtained for such a simple algorithm by establishing a linear relation between the modulated laser power signal and the laser phase in the absence of optical feedback for specific ranges of key optical feedback parameters. We applied CSU on various simulated and experimentally acquired signals using SMI for the retrieval of harmonic and arbitrary displacements and found out that CSU retrieves target displacement with a precision of about λ/10 while consuming much less time and hardware resources. The paper also presents FPGA based hardware design results of CSU and compares its performance with a traditional analytical phase unwrapping method in terms of maximum clock frequency, latency, and on-chip hardware resources. This hardware comparison strongly establishes the advantages of such a fast and computationally light algorithm, readily suitable for large bandwidth, embedded, and real-time sensing applications.