ARAnirban Roy

Anirban Roy

*This is an algorithmically generated author record based on publicly available information.

Anirban Roy has a total of 92 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Development and Validation of a Meal Detection-Based Tool for User Behavior and Meal Management Analysis

    by Andrea Benedetti, John J. Shin, Benyamin Grosman, Ohad Cohen, Robert A. Vigersky, Kamuran Turksoy, Toni L. Cordero, Anirban Roy

    Background: Management of the timing and frequency of meal announcements remains a key determinant of postprandial glycemic control in hybrid automated insulin delivery (AID) systems. Tools that translate routine device data into clinically interpretable insights on meal management behavior for individual patients and broader cohorts remain limited. These tools can potentially provide clinicians and users objective data to formulate personalized meal-related recommendations to safely improve outcomes. Methods: We developed a framework to quantify meal management behavior from routine AID data. The framework combines a meal detection model and a behavioral classification module that categorizes events as on-time, late, or missed based on temporal relationship between detected meals and carbohydrate entries. The meal detection model was trained on data from adults with type 1 diabetes using MiniMed 780G system (NCT04479826) and evaluated in a validation cohort of 157 individuals with type 1 diabetes using advanced hybrid closed-loop (AHCL) (NCT03959423) system. Results: Meal events detection accuracy reached 96.9% in the validation cohort, with precision ≥89.6%, recall ≥93.9%, and area under the curve = 0.996. Application of the tool to the validation cohort showed that a higher proportion of on-time announcements correlated with higher time in range (TIR) (ρ = 0.375, q < 0.001) and lower hyperglycemia. Each 10% increase in on-time announcements corresponded to a 2.66% increase in TIR. Increasing delay of late announcements was associated with worse postprandial glycemic outcomes. Conclusions: This framework enables insights into meal announcement patterns that may support patient counseling, clinical decision support, and large-scale analyses in AHCL systems. Clinical Trial Registration Number: NCT04479826, NCT03959423.

  • A Pilot Study Evaluating Meal Timing, Macronutrient Composition, and Faster-Acting Insulin Aspart Versus Insulin Aspart on Postprandial Glycemia with the Medtronic Advanced Hybrid Closed-Loop System

    by Dale J. Morrison, Melissa H. Lee, Sara Vogrin, Declan Hennessy, Emma Netzer, Hannah M. Jones, Lesley Robinson, Carmel E. Smart, Natalie Kurtz, Anirban Roy, et al.

    Background: This study compared postprandial glycemia with faster-acting insulin aspart (FiAsp) versus insulin aspart delivered by advanced hybrid closed loop (AHCL) according to varying meal composition, time of day, and premeal bolus dose estimation in adults with type 1 diabetes. Materials and Methods: Participants received 11 weeks of insulin aspart and FiAsp in sequence, delivered by the MiniMed™ 780G AHCL algorithm, and within each period ingested 12 standardized meals. Test meals, containing 60 g of carbohydrate but varying by fat and protein content (high fat low protein [HFLP], high fat high protein [HFHP], and low fat low protein [LFLP]), were each eaten in the morning and evening and with a full bolus or 50% reduction. The primary outcome was continuous glucose monitoring time in range (%TIR; 3.9–10.0 mmol/L) 4 h postmeal for FiAsp versus insulin aspart. Results: Twelve participants (mean [standard deviation or SD] age 48 (13) years; 58% male; HbA1c 6.7 (0.7)%/50 (7) mmol/mol) were recruited. When all meal types, timing, and bolus conditions were combined, FiAsp tended to have greater glucose TIR compared with insulin aspart (mean difference = 5.17% [−0.09, 10.43]; P = 0.054), with differences most apparent for morning meals. Insulin formulation did not impact outcomes with full versus 50% premeal bolus or meal composition. HFLP meals resulted in the greatest TIR, followed by HFHP and lowest for LFLP (83.7% [19.6] vs. 74.7% [26.9] vs. 62.0% [23.1], respectively, mean [SD], P < 0.01). Conclusions: For adults with type 1 diabetes who are susceptible to morning hyperglycemia, FiAsp delivered by AHCL and higher fat content of meals may help improve postprandial glycemic outcomes. Clinical Trial Registration: Australian New Zealand Clinical Trials Registry—ACTRN12619001340123.


  • Simplified Meal Management in Adults Using an Advanced Hybrid Closed-Loop System

    by Noga Minsky, Roy Shalit, Andrea Benedetti, Maya Laron-Hirsh, Ohad Cohen, Natalie Kurtz, Anirban Roy, Benyamin Grosman, Amir Tirosh

    Background: The advanced hybrid closed-loop (AHCL) algorithm combines automated basal rates and corrections yet requires meal announcement for optimal performance, which poses a challenge for some. We aimed to compare glucose control in adults with type 1 diabetes (T1D) using the MiniMed TM 780G AHCL system, utilizing simplified meal announcement versus precise carbohydrate (CHO) counting. Methods: In a study involving 14 adults with T1D, we evaluated glycemic control during a 13-week “precise phase,” followed by two 3- to 4-week simplified meal announcement phases: “fixed one-step” (preset of one personalized fixed CHO amount) and “multistep” (entry of multiples of one, two, or three of these presets depending on meal size estimate). Results: The mean age was 45.7 ± 12.4, and 10 participants were male (71%). Mean baseline HbA1c was 6.8% ± 1.2% and time in range (TIR) was 67.5% ± 16.7%. Comparing the fixed one-step to the precise study phase, TIR was similar (75.4 ± 13% vs. 77.7 ± 9%, P = 0.12), and glucose management indicator (GMI) was slightly higher (6.8 ± 0.4 vs. 6.6 ± 0, P = 0.01). Furthermore, there was less level 1 and 2 hypoglycemia (1.6 ± 1% vs. 2.8 ± 2%, P = 0.03 and 0.3 ± 5% vs. 0.65 ± 1%, P = 0.08) but slightly more level 1 and 2 hyperglycemia (17.1 ± 8% vs. 15.0 ± 7%, P = 0.05 and 5.5 ± 5% vs. 3.6 ± 3%, P = 0.04). When comparing the multistep with the precise phase, GMI was identical (6.6%) and TIR superior (80.5 ± 10% vs. 77.7 ± 9%, P = 0.02). Additionally, there was less level 1 hypoglycemia (1.9 ± 1% vs. 2.8 ± 2%, P = 0.01) and a trend for less level 2 hypoglycemia (0.4 ± 0.7% vs. 0.65 ± 1%, P = 0.08). Conclusions: A simplified meal announcement strategy for adults using the MiniMed 780G system, relying on three increments of a fixed one-step CHO amount, may offer a way to improve glycemic control and ease self-care. For patients with more limitations, using one fixed one-step CHO amount could be a safe alternative to meeting most consensus glycemic targets.


  • An Automated Insulin Delivery System with Automatic Meal Bolus Based on a Hand-Gesturing Algorithm

    by Anirban Roy, Benyamin Grosman, Andrea Benedetti, Bahman Engheta, Diana Miller, Maya Laron-Hirsh, Yael Cohen, Roseline Ré, Shannon N., Robert Vigersky, et al.

    Background: Carbohydrate counting (CC) and meal announcements, before eating, introduce a significant burden for individuals managing type 1 diabetes (T1D). An automated insulin delivery system with automatic bolusing that eliminates the need for CC and premeal bolusing (i.e., a hands-free closed-loop [HFCL] system) was assessed in a feasibility trial of adults with T1D. Methods: The system included the MiniMed™ 780G pump and a smartphone-paired smartwatch with the Klue application (Klue, Inc.) that detects eating and drinking gestures. A smartphone algorithm converted gestures into carb amounts that were transmitted to the pump for automatic bolusing. For 5 days, participants ( N = 17, 18–75 years of age) used the system at home with meal announcements based on traditional CC, with the Klue application disabled (Home-stay phase). Thereafter, participants moved to a supervised hotel setting, where the Klue application was enabled for 5 days and meals were not announced (Hotel-stay phase). Participants consumed the same eight test meals (six solid and two liquid) of varying caloric and carb size at the same time and day of the week for both phases, and glycemic metrics were compared. Otherwise, there were no other meal restrictions. Results: The overall time in range (70–180 mg/dL) was 83.4% ± 7.0% and 80.6% ± 6.7% for the Home-stay and Hotel-stay, respectively ( P = 0.08). The average time at 180 mg/dL was 13.5% and 16.3% ( P = 0.1046), respectively. Postprandial glycemia following low-carb test meals was similar between the two phases. The system's ability to accommodate high-carb meals was somewhat limited. There were no episodes of severe hypoglycemia or diabetic ketoacidosis. Conclusion: Preliminary findings show that a HFCL system was safe and maintained overall glycemic control, similar to that observed with traditional CC and manual meal bolusing. By eliminating these daily T1D burdens, a HFCL system may improve quality of life for individuals with T1D. ClinicalTrials.gov number: NCT04964128


  • A Peek Under the Hood: Explaining the MiniMed™ 780G Algorithm with Meal Detection Technology

    by Benyamin Grosman, Anirban Roy, Louis Lintereur, Kamuran Turksoy, Andrea Benedetti, Toni L. Cordero, Robert A. Vigersky, Jennifer McVean, Andrew S. Rhinehart, Ohad Cohen

    The MiniMed™ 780G system (780G) received Conformité Européenne mark in June 2020 and was, recently, approved by the U.S. Food and Drug Administration (April 2023). Clinical trials and real-world analyses have demonstrated MiniMed™ 780G system safety and effectiveness and that glycemic outcomes (i.e., time in range) improve with recommended settings use. In this publication, we will explain the iterative development of the 780G algorithm and how this technology has simplified diabetes management.


  • Open Access

    Safety and Glycemic Outcomes During the MiniMed TM Advanced Hybrid Closed-Loop System Pivotal Trial in Children and Adolescents with Type 1 Diabetes

    by Catherine Pihoker, Dorothy I. Shulman, Gregory P. Forlenza, Kevin B. Kaiserman, Jennifer L. Sherr, James R. Thrasher, Bruce A. Buckingham, Mark S. Kipnes, Bruce W. Bode, Anders L. Carlson, et al.

    Background: During MiniMed™ advanced hybrid closed-loop (AHCL) use by adolescents and adults in the pivotal trial, glycated hemoglobin (A1C) was significantly reduced, time spent in range (TIR) was significantly increased, and there were no episodes of severe hypoglycemia or diabetic ketoacidosis (DKA). The present study investigated the same primary safety and effectiveness endpoints during AHCL use by a younger cohort with type 1 diabetes (T1D). Methods: An intention-to-treat population ( N = 160, aged 7–17 years) with T1D was enrolled in a single-arm study at 13 investigational centers. There was a run-in period (∼25 days) using HCL or sensor-augmented pump with/without predictive low-glucose management, followed by a 3-month study period with AHCL activated at two glucose targets (GTs; 100 and 120 mg/dL) for ∼45 days each. The mean ± standard deviation values of A1C, TIR, mean sensor glucose (SG), coefficient of variation (CV) of SG, time at SG ranges, and insulin delivered between run-in and study were analyzed (Wilcoxon signed-rank test or t -test). Results: Compared with baseline, AHCL use was associated with reduced A1C from 7.9 ± 0.9% ( N = 160) to 7.4 ± 0.7% ( N = 136) ( P < 0.001) and overall TIR increased from the run-in 59.4 ± 11.8% to 70.3 ± 6.5% by end of study ( P < 0.001), without change in CV, time spent below range (TBR) N = 52), an AIT of 2 h ( N = 19) with the 100 mg/dL GT increased mean TIR to 73.4%, reduced TBR 180 mg/dL from 28.7% to 24.4%. During AHCL use, there was no severe hypoglycemia or DKA. Conclusions: In children and adolescents with T1D, MiniMed AHCL system use was safe, A1C was lower, and TIR was increased. The lowest GT and shortest AIT were associated with the highest TIR and lowest TBR and TAR, all of which met consensus-recommended glycemic targets. ClinicalTrials.gov ID: NCT03959423.


  • Unannounced Meal Challenges Using an Advanced Hybrid Closed-Loop System

    by Roy Shalit, Noga Minsky, Maya Laron-Hirsh, Ohad Cohen, Natalie Kurtz, Anirban Roy, Benyamin Grosman, Andrea Benedetti, Amir Tirosh

    Background: The advanced hybrid closed-loop (AHCL) algorithm combines automated basal rates and corrections yet requires meal announcement for optimal outcomes. We aimed to compare the performance of the MiniMed™ 780G AHCL algorithm with and without meal announcement. Methods: In a single-arm study involving 14 adults with type 1 diabetes, we evaluated the safety and efficacy of AHCL when meals were not announced. Participants stayed at a supervised environment for 5 days, during which the outcomes of not announcing meals (≤80 g of carbohydrate) were assessed. Next, participants entered a 90-day at-home “unannounced” phase, during which all meals (≤80 g of carbohydrate) were unannounced, followed by a 90-day at-home phase in which all meals were announced. Results: Time in range (TIR 70–180 mg/dL) was lower in the unannounced versus announced periods (67.5% ± 12.5% vs. 77.7% ± 9.5%; P < 0.01, respectively), with more time spent in hyperglycemia range 180–250 mg/dL (22.7% ± 7.7% vs. 15.7% ± 7.2%) and >250 mg/dL (7.9% ± 6.4% vs. 3.6% ± 2.7%), but less time in hypoglycemia range 54–70 mg/dL (1.6% ± 1% vs. 2.8% ± 1.8%) and 250 mg/dL, and up to 20 g of unannounced carbohydrates did not significantly change the TIR 70–180 mg/dL compared with full announcement. Conclusion: The AHCL system is optimized for use with meal announcement. While not announcing meals of ≤80 g carbohydrates appears to be safe, it results in suboptimal postprandial glycemic control, especially with high-carbohydrate meals. Not announcing small meals (≤20 g carbohydrate) does not deteriorate glycemic control. Clinical Trial Registration number: NCT04479826


  • Improved technology satisfaction and sleep quality with Medtronic MiniMed® Advanced Hybrid Closed-Loop delivery compared to predictive low glucose suspend in people with Type 1 Diabetes in a randomized crossover trial

    by Benjamin J. Wheeler, Olivia J. Collyns, Renee A. Meier, Zara L. Betts, Chris Frampton, Carla M. Frewen, Barbara Galland, Niranjala M. Hewapathirana, Shirley D. Jones, Denis S. H. Chan, et al.

    Background Automated insulin delivery aims to lower treatment burden and improve quality of life as well as glycemic outcomes. Methods We present sub-study data from a dual-center, randomized, open-label, two-sequence crossover study in automated insulin delivery naïve users, comparing Medtronic MiniMed® Advanced Hybrid Closed-Loop (AHCL) to Sensor Augmented Pump therapy with Predictive Low Glucose Management (SAP + PLGM). At the end of each 4-week intervention, impacts on quality of life, sleep and treatment satisfaction were compared using seven age-appropriate validated questionnaires given to patients or caregivers. Results 59/60 people completed the study (mean age 23.3 ± 14.4yrs). Statistically significant differences favoring AHCL were demonstrated in several scales (data shown as mean ± SE). In adults (≥ 18yrs), technology satisfaction favored AHCL over PLGM as shown by a higher score in the DTSQs during AHCL (n = 28) vs SAP + PLGM (n = 29) (30.9 ± 0.7 vs 27.9 ± 0.7, p = 0.004) and DTSQc AHCL (n = 29) vs SAP + PLGM (n = 30) (11.7 ± 0.9 vs 9.2 ± 0.8, p = 0.032). Adolescents (aged 13–17yrs) also showed a higher DTSQc score during AHCL (n = 16) versus SAP + PLGM (n = 15) (14.8 ± 0.7 vs 12.1 ± 0.8, p = 0.024). The DTQ “change” score (n = 59) favored AHCL over SAP + PLGM (3.5 ± 0.0 vs 3.3 ± 0.0, p < 0.001). PSQI was completed in those > 16 years (n = 36) and demonstrated improved sleep quality during AHCL vs SAP + PLGM (4.8 ± 0.3 vs 5.7 ± 0.3, p = 0.048) with a total score > 5 indicating poor quality sleep. Conclusion These data suggest that AHCL compared to SAP + PLGM mode has the potential to increase treatment satisfaction and improve subjective sleep quality in adolescents and adults with T1D.


  • Feasibility Study of a Hybrid Closed-Loop System with Automated Insulin Correction Boluses

    by Revital Nimri, Benyamin Grosman, Anirban Roy, Judith Nir, Naama Fisch Shvalb, Natalie Kurtz, Neta Loewenthal, Michal Gillon-Keren, Ido Muller, Eran Atlas, et al.

    Background: The Medtronic MiniMed™ 670G system adjusts basal insulin delivery in response to continuous glucose monitoring levels and is already in use in clinical practice. We tested the home-based feasibility of the new MiniMed advanced hybrid closed-loop (AHCL) system, which includes several algorithm enhancements and an optional autocorrection bolus mode. Methods: Twelve adolescents and young adults (eight females) with type 1 diabetes [median (interquartile range)] aged 16.6 (15.9, 18.2) years and diabetes duration of 7.1 (4.7, 8.8) years] participated in this single-arm study. The first stage was a 6-day open-loop run-in period, with the predictive low-glucose suspend feature on. This was followed by 6 days/5 nights in a supervised hotel setting, using the AHCL system, including closed-loop challenges (missed meal bolus, late meal bolus, and physical activity); and finally, 3 weeks with unrestricted home use. Glycemic parameters were compared between the open-loop and closed-loop periods. Results: Participants spent 93.3% (4.7) of the time in SmartGuard™ Auto Mode. Hemoglobin A1C levels decreased from median (interquartile range) 7.1% (6.7, 7.9) at baseline to 6.8% (6.6, 7.4) at study end, after 4 weeks ( P = 0.0027). Time in range (TIR) (70–180 mg/dL) was 68.4% (10.6) and time below 70 mg/dL was 4% (3.5) during open-loop; and 74% (6.1) and 2.6% (1.9), respectively, during the closed-loop at home phase ( P = 0.06, P = 0.27). TIR increased during the nighttime, from 64.6% (17.4) to 80.7% (7.8), P = 0.007, without change in time below 70 mg/dL ( P = 0.15). No serious adverse events occurred. Conclusions: The new AHCL system demonstrated safety and effectiveness in controlling day and night glucose levels.


  • 1006-P: Personalized Hybrid Closed-Loop Therapy Using a Digital Twin in Patients with Type 1 Diabetes: At-Home Data

    by Benyamin Grosman, Anirban Roy, DI Wu, Neha Parikh, Louis J. Lintereur, Nicole Schneider, Ronald L. Brazg, Satish K. Garg, Robert Vigersky

    Background: MiniMed™ 670G system user-dependent settings influence glycemic outcomes in persons with type 1 diabetes. The ability of a mathematical model-based digital-twin program to personalize and automate user-dependent MiniMed™ 670G settings was assessed. Method: A small feasibility study recruited experienced MiniMed™ 670G system users (N=19 users, >14yrs) to complete an at-home 3-weeks run-in phase with Auto Mode enabled followed by a 2-weeks study phase. Run-in data were used to generate a digital twin that personalized system settings during the study phase that comprised a non-randomized Auto Mode and Manual Mode arm. System use and glycemic outcomes were assessed. Results: Personalized Auto Mode (Auto Mode+) reduced the mean(SD) of sensor glucose (SG) from 151.1(12.3)mg/dL to 149.3(11.6)mg/dL and increased time spent in target SG range from 74.0% to 75.9% (Table). Personalized Manual Mode (Manual Mode+) demonstrated a mean SG and TIR of 150.1mg/dL and 67.4%, respectively. Conclusions: Improved Auto Mode therapy and acceptable Manual Mode (sensor-augmented pump) therapy were observed utilizing a digital-twin program to personalize MiniMed™ 670G system settings in subjects whose glycemic control was excellent at baseline. This is the first demonstration that a digital-twin can successfully automate and personalize insulin pump therapy. Disclosure: B. Grosman: Employee; Self; Medtronic. A. Roy: Employee; Self; Medtronic. D. Wu: None. N. Parikh: None. L.J. Lintereur: Employee; Self; Medtronic. N. Schneider: None. R.L. Brazg: None. S.K. Garg: Advisory Panel; Self; Boehringer Ingelheim Pharmaceuticals, Inc., Eli Lilly and Company, Medtronic, Novo Nordisk A/S, Roche Diagnostics France. Research Support; Self; Dexcom, Inc., Eli Lilly and Company, Medtronic. R. Vigersky: Employee; Self; Medtronic.