Blood Count on a Smartphone Microscope: Challenges Uma P. Moravapalle Georgia Tech parthavi@gatech.edu Amit Deshpande Microsoft Research amitdesh@microsoft.com Ashish Kapoor Microsoft Research akapoor@microsoft.com Ramachandran Ramjee Microsoft Research ramjee@microsoft.com Priya Ravi Atreya Clinic rampriya2000@yahoo.com ABSTRACT Low-cost lenses with magnifications of 150-200x are being sold in the market today as accessories for mobile smart- phones. Attaching these lenses to a smartphone camera cre- ates low-cost, ultra-portable digital microscopes, with a po- tential for significant impact on applications in a variety of fields such as healthcare, agriculture, education etc. In this paper, we consider a low-cost do-it-yourself Com- plete Blood Count (CBC) application using a smartphone microscope. We discuss several unique challenges that come up in implementing this application that include preparing the blood sample, correcting the small field of view and blur of the lens, and automating the cell counting procedure. We present our approach to overcome these challenges and re- port early promising results on counting red blood cells. CCS Concepts •Applied computing → Health informatics; 1. INTRODUCTION The number of smartphones worldwide is approaching 2 billion today [6]. The mobile accessory market, riding on the coattails of the growth in smartphones, is expected to grow to $100 billion by 2020 [5]. One such accessory, the smartphone camera microscope lens, has made tremendous progress recently in both cost and functionality. For ex- ample, in the commerical market today, one can obtain smartphone lens attachments with 150-200X magnification for $15-$30 [4, 11]. Further, the Foldscope project promises a lens with up to 2000X magnification for less than a dol- lar [2]. These advances in smartphone microscopy can open up a variety of applications with significant impact in healthcare, agriculture, education, etc. In this paper, we focus on chal- lenges in implementing one such application, namely, blood cell counting using a smartphone microscope. Human blood is composed of blood cells or corpuscles sus- pended in blood plasma. Plasma constitutes about 55% of Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full cita- tion on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or re- publish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. HotMobile ’17, February 21-22, 2017, Sonoma, CA, USA c  2017 ACM. ISBN 978-1-4503-4907-9/17/02. . . $15.00 DOI: http://dx.doi.org/10.1145/3032970.3032986 the volume and is mostly (up to 95% by volume) water. There are three kinds of blood cells: (a) red blood cells (RBCs) or erythrocytes, (b) white blood cells (WBCs) or leukolytes, (c) platelets or thrombocytes. Complete blood count (CBC) test is one of the most widely ordered laboratory tests. A complete blood count (CBC) test report helps confirm the diagnoses for several medical conditions, for example, anemia (low RBC count), leukemia (low RBC count), dehydration (high RBC count), renal cell carcinoma (high RBC count), bone marrow failure (low WBC count), lupus (low WBC count), stress (high WBC count), infection (high WBC count), and dengue (low platelet count). As we discuss in Section 2, CBC can be performed either by manual count using a microscope or through an auto- mated flow cytometry machine. However, the manual count is tedious and can be error-prone. Thus, most diagnostic labs today use the automated approach which increases cost and also requires a visit to the diagnostic lab. In this paper, we investigate the feasibility of a do-it-yourself blood test that automates the manual count approach using a smart- phone microscope. This allows the CBC test to be done anywhere and at low cost, thereby addressing the needs of users in both developed and developing nations. While smartphone lens attachments are inexpensive and provide good magnification, a key drawback of these lenses is their small field-of-view. For example, the practical field-of- view of one of the lenses [4] is only an area of about 0.2 mm× 0.2 mm while manual blood cell counting protocol requires observation of an area of few square mm – an area that is up to two orders of magnitude larger. This introduces a number of challenges in blood sample imaging and the vision algorithms used in their processing. We describe these and other challenges that arise in handling blood in Section 4. We discuss our approach for addressing these challenges and our preliminary results in computing red blood cell count in Sections 5 and 6. As part of future work, we plan to enhance our processing to support other blood cell types as well as scale our evaluation. If successful, we believe that a low-cost do-it-yourself CBC test using a smartphone microscope can complement conventional lab blood testing in situations where convenience or cost considerations pre- vent/delay users from visiting a conventional lab. 2. BACKGROUND In this section, we describe the essential parameters mea- sured by a CBC test and the laboratory procedure for it. The complete blood count (CBC) test procedure starts with a phlebotomist drawing approximately 0.5 ml of the