Abstract—A novel method of validation of the mathematical model for batteries that power Medtronic’s Implantable Cardioverter Defibrillators (ICDs) is presented. In a conventional approach used in the past, the model has been validated against data collected in controlled laboratory conditions. To supplement this approach, we now validate the model against ICD performance data reported from devices used in the field for periods ranging from about five to seven years. The key model output is ICD “charge time” – the time required to charge a high voltage capacitor in preparation to deliver shock to the heart. This validation is carried out for five of Medtronic’s ICD designs and very close agreement is obtained between model predictions and field data. I. BACKGROUND EDTRONIC has developed lithium primary batteries with silver vanadium oxide cathodes as power sources for implantable cardioverter defibrillators (ICD’s).[1] The ICD effectively treats multiple disease states of the heart – for example, when the heart beats slower than normal it provides pacing with low energy pulses; when the heart beats faster than normal it provides more frequent low energy pulses. Most critically, the ICD treats ventricular fibrillation. Also known as sudden cardiac death, ventricular fibrillation is a generally fatal condition, characterized by rapid, erratic contraction of the heart resulting in little or no pumping of blood. Within seconds of detecting ventricular fibrillation, the ICD delivers a high- energy pulse (typically up to 35 J) to the heart to bring it back to normal rhythm. To deliver this life-saving therapy, the ICD battery charges a capacitor to the desired energy level in as short a time as possible, and the capacitor is subsequently discharged through the heart. Because prompt therapy is desirable, the capacitor charge-time, typically in the range of 5 to 15 s, is a key measure of device performance. Our group has earlier developed a mathematical model to predict ICD charge-times over battery life.[1] The model has been validated for all of Medtronic’s ICD battery designs under rigorous laboratory test conditions (e.g., specific applied load and energy delivered during pulsing).[1] In this work, we conduct an independent Manuscript received June 19, 2009. Medtronic Energy and Component Center, Brooklyn Center, MN 55311 USA (phone: 763-514-1071; fax: 763-514-1177; e-mail: partha.m.gomadam@medtronic.com). validation of the model by comparing its charge-time predictions against data observed from ICDs functioning in the field. II. THE MODEL The mathematical model developed for ICD batteries is described in detail in Ref. [1]. The model predicts the background voltage (V b ), the DC resistance (R DC ), and the charge-time (T charge ) as functions of delivered capacity (q del ) and time (t) since the device has been implanted. The equations used to predict the background voltage and DC resistance can be represented conceptually in the form of transfer functions: ( ) b b del V V q = (1) and ( ) , DC DC del R R q t = (2) The dependence of the DC resistance, not just on capacity, but also on time is an important feature that enables accurate model predictions, especially in the latter part of battery discharge. Based on the quantities calculated above, the model calculates charge-time over the life of the ICD: ( ) 2 2 1 , load DC charge b load charge del R E R T V R T q t    = +       = (3) where E is the energy delivered by the battery during the pulse and R load is the resistance of the load under pulse. For specified therapy energy, therapy frequency and load resistance, the model predicts charge-time as a function of delivered capacity and the time of use. These quantities are readily measurable under specified test conditions, when the charge-time predictions have been validated against data for all of Medtronic’s ICD battery designs. For an exemplary design, Fig. 1 shows the close agreement between the predicted and observed charge-times of several years of Predicting Charge-Times of Implantable Cardioverter Defibrillators Parthasarathy M. Gomadam, Jason R. Brown, Erik R. Scott, and Craig L. Schmidt M 3020 31st Annual International Conference of the IEEE EMBS Minneapolis, Minnesota, USA, September 2-6, 2009 978-1-4244-3296-7/09/$25.00 ©2009 IEEE Authorized licensed use limited to: Medtronic Library. Downloaded on November 19, 2009 at 17:27 from IEEE Xplore. Restrictions apply.