An inexpensive smartphone-based device for point-of-care ovulation testing

Potluri V; Kathiresan PS; Kandula H; Thirumalaraju P; Kanakasabapathy MK; Kota Sai Pavan S; Yarravarapu D; Soundararajan A; Baskar K; Gupta R; Gudipati N; Petrozza JC; Shafiee H

doi:10.1039/c8lc00792f

An inexpensive smartphone-based device for point-of-care ovulation testing

Potluri V, Kathiresan PS, Kandula H, Thirumalaraju P, Kanakasabapathy MK, Kota Sai Pavan S, Yarravarapu D, Soundararajan A, Baskar K, Gupta R, Gudipati N, Petrozza JC, Shafiee H

Published December 18, 2018 Lab on a Chip, 19(1), 59-67

DOI 10.1039/c8lc00792f PMID 30534677 PMC PMC6321627

RRM Academy Synopsis

AI Smartphone Device Detects Ovulation From Dried Saliva

In a study of six women, researchers built a smartphone attachment that uses AI to spot fern-like patterns in dried saliva and flag ovulation. Each woman was tracked for one cycle. The device identified ovulation with 99.5% accuracy in under 31 seconds, with no internet needed.

Key Findings

Testing human saliva from six women, the AI system identified ovulation with 99.5% accuracy (CI: 97.25% to 99.99%), with 99.01% sensitivity and 100% specificity.
Using artificial saliva made to mimic ferning patterns, the system reached 90% accuracy (95% CI: 84.98% to 93.78%), with 97.62% sensitivity and 84.48% specificity.
Each saliva sample took under 31 seconds to analyze and needed no internet connection, since the AI ran locally on the phone.
The smartphone attachment and microfluidic device cost about $14 in materials and were built to be reused, versus an estimated $35 to $40 per cycle for disposable urine ovulation tests.
In a separate test using artificial saliva samples, the trained algorithm performed consistently across five different smartphone models, with no extra training needed for each device.

Interpretation

This device-validation study followed six healthy women, aged 20 to 22, across one menstrual cycle, confirming ovulation timing with a standard urine LH and estrogen test. Accuracy was higher with human saliva than with artificial saliva. The authors note the system may work less well in women with hormone imbalances, ovarian cysts, or those taking fertility drugs. The findings support larger studies of this device.

RRM Context

Saliva ferning is one of several body signs used by fertility awareness-based methods to find the fertile window, along with basal body temperature, cervical mucus, and urine hormone changes. Reading ferning by eye varies from person to person. A device that reads it the same way each time is designed to remove that variation. A low-cost, reusable, offline tool like this could also cut the day-to-day cost of fertility tracking, compared with disposable hormone test strips.

Our editorial summary of this paper, not the article's abstract.

Abstract

The ability to accurately predict ovulation at-home using low-cost point-of-care diagnostics can be of significant help for couples who prefer natural family planning. Detecting ovulation-specific hormones in urine samples and monitoring basal body temperature are the current commonly home-based methods used for ovulation detection; however, these methods, relatively, are expensive for prolonged use and the results are difficult to comprehend. Here, we report a smartphone-based point-of-care device for automated ovulation testing using artificial intelligence (AI) by detecting fern patterns in a small volume (<100 μL) of saliva that is air-dried on a microfluidic device. We evaluated the performance of the device using artificial saliva and human saliva samples and observed that the device showed >99% accuracy in effectively predicting ovulation.

Topics

Fertility Awareness › Technology › Home Hormone Tests · Menstrual Cycle › Cycle Biomarkers › Cervical Mucus · Diagnostics › Cycle Biomarkers › Ovulation Detection

Prudhvi Thirumalaraju, Karthik Baskar, Raghav Gupta, John C Petrozza, Neeraj Gudipati, Manoj Kumar Kanakasabapathy, Hemanth Kandula, Preethi Sangeetha Kathiresan, Sandeep Kota Sai Pavan, Vaishnavi Potluri, Hadi Shafiee, Anand Soundararajan, Divyank Yarravarapu

P Thirumalaraju, K Baskar, R Gupta, Jack Petrozza, J Petrozza, N Gudipati, M Kanakasabapathy, H Kandula, P Kathiresan, S Pavan, V Potluri, H Shafiee, A Soundararajan, D Yarravarapu

PMID 30534677 30534677 DOI 10.1039/c8lc00792f 10.1039/c8lc00792f Potluri et al. 2018, Potluri 2018

An inexpensive smartphone-based device for point-of-care ovulation testing

AI Smartphone Device Detects Ovulation From Dried Saliva

Key Findings

Interpretation

RRM Context

Abstract

Topics

By this author

Reproductive surgery: revisiting its origins and role in the modern management of fertility

Related research

Helping Patients to Predict and Confirm Ovulation with the Use of Combined Urinary Hormonal and Smartphone Technology: A Proof-of-Concept Retrospective Descriptive Case Series

Efficacy of methods for determining ovulation in a natural family planning program

Detection of ovulation, a review of currently available methods

Evaluation of the Lady Free Biotester in determining the fertile period