The Doctor of Philosophy Seminar and Examination of Christina Pacholec

Biography

Christina Pacholec is a Ph.D. candidate in Biomedical and Veterinary Sciences at Virginia Tech. Prior to her Ph.D. she received her bachelor’s in biology from Iowa State University followed by a Doctor of Veterinary Medicine from Kansas State University in 2017. She went on to complete a rotating internship at Auburn University followed by a veterinary clinical pathology residency at Virginia Tech. Her research combines veterinary clinical pathology with artificial intelligence to create new diagnostics for veterinary patients. Additionally, she has a strong interest in and focus on quality assurance, exploring the differences between the quality assurance needs in veterinary clinical pathology and artificial intelligence systems, and their implementation in a clinical setting.

Funded by

• Veterinary Memorial Fund
• VMCVM Office of Research and Graduate Studies
• American Kennel Club Canine Health Foundation
• Morris Animal Foundation
• ImpriMed

Awards and Academic Achievements

• David-Thompson Foundation Award 2024

Lay Language Abstract 

Recent advances in technology have led to rapid growth in artificial intelligence. This expansion has naturally led to new diagnostic tests in both the medical and veterinary medical fields. In veterinary medicine, before diagnostic tests are offered to patients, they undergo rigorous testing to ensure they are safe, reliable, and trustworthy. For standard diagnostic tests, the American Society of Veterinary Clinical Pathology (ASVCP) has published guidelines that make recommendations on producing and maintaining new diagnostics. However, these guidelines do not cover the unique needs of artificial intelligence systems.

In the present work, we review the current literature to provide the best recommendations on producing and maintaining artificial intelligence systems for use in veterinary pathology. This work introduces veterinary pathologists to basic concepts of artificial intelligence-based diagnostics and provides the minimal quality requirements for artificial intelligence systems in the medical field. By providing high-quality diagnostics and standardization of quality assurance, we maintain trust and reliability in the diagnostic tests we, as a profession, offer. This allows us to better serve our patients, clients, and community while advancing veterinary medicine in a way that benefits all. Despite rapid advancements in artificial intelligence-based diagnostics, little is known about some basic requirements for building these systems. Therefore, to fill some gaps, this work explores the ideal magnification, image type (color versus grey-scale) and number of images needed to build an artificial intelligence system. The findings of this research suggest that higher magnification with either color or grey-scale images is ideal for building a convolutional neural network (type of artificial intelligence). Additionally, the ideal number of images to use for a two-class problem is 150 images per class. This work is foundational in understanding the requirements of convolutional neural networks and allows for future studies.

Publications

Pacholec C, Flatland B, Xie H, Zimmerman K. Harnessing artificial intelligence for enhanced veterinary diagnostics: A look to quality assurance, Part I Model development. Veterinary clinical pathology. 2024.

Pacholec C, Flatland B, Xie H, Zimmerman K. “Harnessing Artificial Intelligence for Enhanced Veterinary Diagnostics: A Look to Quality Assurance, Part II External Validation.” Veterinary clinical pathology. 2024.

Pacholec C, Xie H, Curnin J, Lin A, Zimmerman K. “Impact of Magnification, Image Type, and Count on Convolutional NeuralNetwork Performance in Differentiating Canine Large Cell Lymphoma from Nonlymphoma via Lymph Node Cytology”. Veterinary Clinical Pathology. (submitted 2024, pending)

Xu X, Lin Y, Yin L, Serpa PD, Conacher B, Pacholec C, Carvallo F, Hrubec T, Farris S, Zimmerman K, Wang X. Spatial Transcriptomics and Single-Nucleus Multi-omics Analysis Revealing the Impact of High Maternal Folic Acid Supplementation on Offspring Brain Development. Nutrients. 2024 Nov 7;16(22):3820.

Pacholec C, Carvallo F, LeCuyer TE, Todd MS, Ramirez-Barrios R, Weisman J, Zimmerman K. “What’s your diagnosis? Cecal smear in a peafowl. Veterinary Clinical Pathology (2023)

Johnson Z, Xiguang X, Pacholec C and Hehuang X. “Systematic Evaluation of Parameters in RNA Bisulfite Sequencing Data Generation and Analysis.” NAR Genomics and Bioinformatics 4, no.2 (2022): iqac045.

Pacholec C, Lisciandro GR, Masseau I, Donnelly L, Declue AE, and Reinero CR. “Lung Ultrasound Nodule Sign for Detection of Pulmonary Nodule Lesions in Dogs: Comparison to Thoracic Radiography Using Computed Tomography as the Criterion Standard.” The Veterinary Journal 275 (2021): 105727.

Jaffey JA, Matheson J, Shumway K, Pacholec C, Ullal T, Van den Bossche L, Fieten H, Ringold R, Lee KJ, and DeClue AE. "Serum 25-hydroxyvitamin D concentrations in dogs with gallbladder mucocele." Plos one 15, no. 12 (2020): e0244102.

Pacholec C and Cohn LA. “Cytauxzoonosis in Cats.” Todays Veterinary Practice (TVP). 2020

Presentations

Invited speaker

Fall 2025, The use of Artificial Intelligence in Veterinary Clinical Pathology. European College of Internal Medicine

Fall 2024, The use of Artificial Intelligence in Veterinary Clinical Pathology. American College of Veterinary Pathology Annual Conference

Platform presentations

2022, Mystery case 2: Fluid from the ventral neck mass in a Guinea pig. American College of Veterinary Pathology Annual Conference

2022, Histoplasmosis in feline blood smear. Southeastern Veterinary Pathology Conference

Poster presentations

Fall 2024, Impact of Magnification, Image Type, and Count on Convolutional Neural Network Performance in Differentiating Canine Large Cell Lymphoma from Nonlymphoma via Lymph Node Cytology. American College of Veterinary Pathology

Annual Conference

2024, Use of Artificial Intelligence for Detection of Minimal Residual Disease in Dogs Being Treated for Lymphoma. Virginia Tech Biomedical Sciences and Pathobiology Research Symposium

2023, Effects of Image Magnification, Type and Resolution on the Accuracy of an Artificial Intelligence Model. Virginia Tech Biomedical Sciences and Pathobiology Research Symposium

Examination Graduate Committee

Major Advisor/Chair:

Kurt Zimmerman, Chair, DVM, Ph.D., DACVP (CP/AP)
Professor
Department of Biomedical Sciences and Pathobiology

Graduate Advising Committee Members:

Hehuang (David) Xie, Co-chair, DVM, Ph.D., DACVP (AP)
Professor
Department of Biomedical Sciences and Pathobiology

Chandra Saravanan, DVM, Ph.D., DACVP (AP)
Novartis

Kevin Lahmers, DVM, Ph.D., DACVP (AP)
Professor
Department of Biomedical Sciences and Pathobiology