Rameau Lab

Current Projects

We developed the Laryngology Innovation Laboratory at Weill Cornell Medical College with the intent to disrupt and innovate the fields of voice, deglutition and airway science, via transdisciplinary collaborations with engineers, bio-acousticians, designers and computer scientists. Since our start in 2018, we have a growing contingent of engineering and medical students joining us every semester from institutions across the Tri-State Area. Our group has received three institutional grants, two foundation grants, and two NCATS grants over the past 4 years. 

A Deep Learning Pipeline for Automated Classification of Pathology in Laryngoscopy Videos

This interdisciplinary collaboration between laryngologists and computer scientists aims to use deep learning to automate both dataset creation and disease classification in video-laryngoscopy. Using deep learning to classify pathology on laryngoscopy images involves two steps: (1) building a labeled dataset and (2) using the dataset to train a classifier to distinguish healthy from pathologic images. A deep learning model requires a large amount of training data, and building a dataset by manually extracting informative frames from laryngoscopy videos is a research bottleneck. We have developed a deep learning classifier capable of automatically identifying key frames from laryngoscopic videos.
A flow chart showing Laryngoscopy Video frames on the left flowing into two distinct images of Informative Frames vs Uninformative Frames on the right.

Achievements:

  • Our deep learning classifier of key frames has 87% precision, demonstrating the potential to facilitate accurate and efficient creation of deep learning datasets across a range of laryngeal diseases. As a pilot study, we used the informative frame classifier to create a dataset to train a model to recognize vocal fold polyps, which achieved a classification accuracy of 77.4%.

  • The continuation of this work was awarded the 2021 American Laryngological Voice and Research Education Grant. 

Vacuum Helmet to Contain Pathogen-Bearing Droplets in Dental, Otolaryngological and Ophthlamologic Outpatient Interventions.

Clinic encounters of dentists, otolaryngologists and ophthalmologists inherently exposes these specialists to enhanced risk of SARS-CoV-2 infection, thus threatening them, their patients and their practices. Here we propose to computationally model and experimentally test a novel aerosol elimination helmet device that would act as an additional primary barrier to infection, while maintaining open access to the patient’s face. Our device would eliminate the need for negative-pressure aeration in clinic rooms, an often economically and structurally impractical option. We will evaluate the effectiveness of the proposed device by simulating dynamics of ejected droplets during a violent expiratory event. The results of these predictive simulations will be employed to optimize and tailor our design based on the patient’s condition or the specific intended use of the device, before experimentally testing several prototypes of the envisioned vacuum helmet in healthy volunteers performing cough and sneeze. Our preliminary results indicate the effectiveness of the proposed design in eliminating over 98% of pathogen-bearing aerosols and droplets generated during most extreme expiratory events. 
Figure 1: The vacuum helmet proposed design. Figure 2. Airflow simulation in the upper airways using computational fluid dynamics. Figure 3. The departure of droplets of different sizes from the mouth during the most extreme expiratory events when the vac

Figure 1: The vacuum helmet proposed design. Figure 2. Airflow simulation in the upper airways using computational fluid dynamics. Figure 3. The departure of droplets of different sizes from the mouth during the most extreme expiratory events when the vacuum helmet is used. Difference between intervention and reference counts for positive controls stratified by particle size. Positive values represent higher intervention counts relative to background; negative values represent lower intervention counts.

Achievements
COVID19 Sustainability Rapid Response Seed Fund Recipient, Cornell Atkinson Center for Sustainability - Co-PI with Prof. Mahdi Esmaily, Sibley School of Mechanical and Aerospace Engineering 

Is Office Laryngoscopy an Aerosol-Generating Procedure?

The novel coronavirus, SARS-CoV-2, has highlighted a multitude of deficiencies in our healthcare system and scientific knowledge. One of the most disconcerting of these has been our incomplete understanding of aerosol-generating procedures (AGPs). Procedures and interventions involving the upper and lower airways have been linked to increased infection rates among healthcare professionals. Although a number of studies have investigated the aerosol-generating potential of specific procedures, evidence has been ambiguous or contradictory. Office laryngoscopy is a prime example, being a source of concern, debate, and speculation during this crisis. In this study, we quantify aerosolization associated with flexible and rigid office laryngoscopy, and critically review methods for measuring the aerosolizing potential of medical and surgical interventions.

Left:Experimental setup for flexible laryngoscopy and Right: rigid laryngoscopy

Achievements:

  • Rameau A, Lee M, Enver N, Sulica L. Is Office Laryngoscopy an Aerosol Generating Procedure? Accepted for publication: July 1st, 2020. Laryngoscope (expedited, no revision needed acceptance)

  • Our study suggests that office laryngoscopy may not be an AGP according to the standard medical definition, namely aerosol production more than normally expected from breathing, coughing, sneezing, and speaking.

An Open-Source 3D Printed Laryngeal Model for Injection Laryngoplasty Training

Simulators are becoming increasingly important in surgical training, in part due to contemporary educational trends such as resident work hour restrictions and an increasing emphasis on competency-based training. Laryngology procedures in particular—especially those performed in the office—present unique training challenges. Percutaneous injection laryngoplasty, for example, is increasingly performed in the outpatient setting on awake patients, a less optimal teaching environment for residents and fellows compared to the operating room. We have designed an adult model injection laryngoplasty trainer using 3D-printing technology with emphasis on affordability, user friendliness, open-source accessibility, and surgical fidelity. 
Laryngoplasty Training

Novel delivery method of transgender voice therapy using a mobile application

Transgender patients’ voices are closely related to their gender identity. Social isolation and emotional distress can result when patients’ voices are inconsistent with their self-identified gender. Transgender patients’ voices can be rehabilitated to improve congruence with their gender identity through behavioral interventions such as voice therapy. However, access to voice therapy is a significant struggle due to insurance, financial, social and cultural concerns. The development of easily accessible alternatives for behavioral voice modification in transgender patients is crucial to improving their voice-related quality of life. This study addresses this deficit in access to these health services through the development of a mobile application to deliver transgender voice therapy

Attuned Voice Therapy App Logo

Transgender patients’ voices are closely related to their gender identity. Social isolation and emotional distress can result when patients’ voices are inconsistent with their self-identified gender. Transgender patients’ voicesAttuned can be rehabilitated to improve congruence with their gender identity through behavioral interventions such as voice therapy. However, access to voice therapy is a significant struggle due to insurance, financial, social and cultural concerns. The development of easily accessible alternatives for behavioral voice modification in transgender patients is crucial to improving their voice-related quality of life. This study addresses this deficit in access to these health services through the development of a mobile application to deliver transgender voice therapy.

Achievements:

  • 2019-2021 Dean’s Diversity and Health Disparity Research Award, Weill Cornell Medicine

  • We have nearly completed the app development via a multidisciplinary team, including laryngologists, speech pathologists, transgender individuals, programmers and designers. 

MyophonX

Achievements:

• Rameau A. Pilot study for a novel and personalized voice restoration device for patients with laryngectomy. Head Neck. 2020;42(5):839-845.

• New International Application No. PCT/US2020/034917, Based on U.S. Provisional App. No. 62/854,128, submitted June 2020

• Winner, 2018 Annual Weill Cornell Medicine Health Innovation Hackathon

• This work was supported by a grant (UL1-TR-002384) from the Clinical and Translational Science Center at Weill Cornell Medicine, which is funded by the National Center for Advancing Translational Sciences (NCATS)

• Project selected for start-up acceleration by ElabNYC, a competitive 6-month program launched by the City of New York, designed for NY scientists interested in forming new ventures in the life sciences and healthcare technology sectors.

• One of the four Cornell start-ups selected for podium presentation at the Cornell Entrepreneurship Summit in November 2019 at the Times Building, New York, NY. 

Weill Cornell Medicine Rameau Lab 240 East 59th Street New York, NY 10022 Phone: (646) 962-7464