Back to all papers

Utilizing a publicly accessible automated machine learning platform to enable diagnosis before tumor surgery.

Authors

Hosseinzadeh F,Liu G,Tsai E,Mahmoudi A,Yang A,Kim D,Fieux M,Levi L,Abdul-Hadi S,Adappa ND,Alt JA,Altartoor KA,Banyi N,Challa M,Chandra R,Chang MT,Chen PG,Cho DY,de Choudens CR,Chowdhury N,Colon CM,DelGaudio JM,Del Signore A,Dorismond C,Dutra D,Edalati S,Edwards TS,Ferriol JB,Geltzeiler M,Georgalas C,Govindaraj S,Grayson JW,Gudis DA,Harvey RJ,Heffernan A,Hwang PH,Iloreta AM,Knight ND,Kohanski MA,Lerner DK,Leventi A,Lee LH,Lubner R,Mahomva C,Massey C,McCoul ED,Nayak JV,Pak-Harvey E,Palmer JN,Pandrangi VC,Psaltis AJ,Raviv J,Sacks P,Sacks R,Schaberg M,Soudry E,Sweis A,Thamboo A,Turner JH,Wang SX,Wise SK,Woodworth BA,Wormald PJ,Patel ZM

Affiliations (25)

  • Department of Otolaryngology-Head & Neck Surgery, Stanford University School of Medicine, Stanford, CA, USA.
  • Hospices Civils de Lyon, Centre Hospitalier Lyon Sud, Service d'ORL, d'otoneurochirurgie et de chirurgie cervico-faciale, France, Pierre Bénite cedex F-69495; Université de Lyon, Université Lyon 1, Lyon, France.
  • Department of Otolaryngology-Head & Neck Surgery, University of Puerto Rico School of Medicine, San Juan, PR, USA.
  • Department of Otolaryngology-Head and Neck Surgery, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
  • Department of Otolaryngology-Head and Neck Surgery, University of Utah, Salt Lake City, UT, USA.
  • Department of Otolaryngology-Head and Neck Surgery, Emory University School of Medicine, Atlanta, GA, USA.
  • Department of Surgery, St. Paul's Sinus Centre, University of British Columbia, Vancouver, BC, Canada.
  • Department of Otolaryngology-Head and Neck Surgery, University of Texas Health Science Center, San Antonio, TX, USA.
  • Department of Otolaryngology-Head and Neck Surgery, Vanderbilt University School of Medicine, Nashville, TN, USA.
  • Department of Otolaryngology-Head and Neck Surgery, University of Alabama at Birmingham School of Medicine, Birmingham, AL, USA.
  • Department of Otolaryngology-Head and Neck Surgery, Mount Sinai Icahn School of Medicine, New York, NY, USA.
  • Department of Otolaryngology-Head and Neck Surgery, Central Adelaide Local Head Network and University of Adelaide, Adelaide, Australia.
  • Department of Otolaryngology-Head and Neck Surgery, Oregon Health & Science University, Portland, OR, USA.
  • Medical School, University of Nicosia, 93 Agiou Nikolaou Street, Engomi, Nicosia, Cyprus.
  • Department of Otolaryngology-Head and Neck Surgery, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, USA.
  • Macquarie University, Medicine and Health Sciences, Sydney, Australia.
  • Rhinology and Skull Base Research Group, St Vincent's Centre for Applied Medical Research, University of New South Wales, Sydney, NSW, Australia.
  • Department of Pathology and Laboratory Medicine, St. Paul's Hospital, University of British Columbia, Vancouver, BC, Canada.
  • Department of Surgery, University of South Dakota, Sioux Falls, SD, USA.
  • Department of Otorhinolaryngology, Ochsner Health, New Orleans, LA, USA.
  • Endeavor Health, Evanstown, IL, USA.
  • University of Sydney, Sydney, Australia.
  • Department of Otolaryngology Head and Neck Surgery, Rabin Medical Center, Petah Tikva, Israel.
  • Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel.
  • Department of Otolaryngology-Head & Neck Surgery, Stanford University School of Medicine, Stanford, CA, USA. [email protected].

Abstract

In benign tumors with potential for malignant transformation, sampling error during pre-operative biopsy can significantly change patient counseling and surgical planning. Sinonasal inverted papilloma (IP) is the most common benign soft tissue tumor of the sinuses, yet it can undergo malignant transformation to squamous cell carcinoma (IP-SCC), for which the planned surgery could be drastically different. Artificial intelligence (AI) could potentially help with this diagnostic challenge. CT images from 19 institutions were used to train the Google Cloud Vertex AI platform to distinguish between IP and IP-SCC. The model was evaluated on a holdout test dataset of images from patients whose data were not used for training or validation. Performance metrics of area under the curve (AUC), sensitivity, specificity, accuracy, and F1 were used to assess the model. Here we show CT image data from 958 patients and 41099 individual images that were labeled to train and validate the deep learning image classification model. The model demonstrated a 95.8 % sensitivity in correctly identifying IP-SCC cases from IP, while specificity was robust at 99.7 %. Overall, the model achieved an accuracy of 99.1%. A deep automated machine learning model, created from a publicly available artificial intelligence tool, using pre-operative CT imaging alone, identified malignant transformation of inverted papilloma with excellent accuracy.

Topics

Journal Article

Ready to Sharpen Your Edge?

Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.

We respect your privacy. Unsubscribe at any time.