Artificial Intelligence in the Diagnosis of Invasive Mold Infection: Development of an Automated Histologic Identification System to Distinguish Between <i>Aspergillus</i> and Mucorales

It is important to realize that guidelines cannot always account for individual variation among patients. They are not intended to supplant physician judgment with respect to particular patients or special clinical situations. IDSA considers adherence to these guidelines to be voluntary, with the ultimate determination regarding their application to be made by the physician in the light of each patient's individual circumstances.

Fungal infections are becoming more frequent because of expansion of at-risk populations and the use of treatment modalities that permit longer survival of these patients. Because histopathologic examination of tissues detects fungal invasion of tissues and vessels as well as the host reaction to the fungus, it is and will remain an important tool to define the diagnostic significance of positive culture isolates or results from PCR testing. However, there are very few instances where the morphological characteristics of fungi are specific. Therefore, histopathologic diagnosis should be primarily descriptive of the fungus and should include the presence or absence of tissue invasion and the host reaction to the infection. The pathology report should also include a comment stating the most frequent fungi associated with that morphology as well as other possible fungi and parasites that should be considered in the differential diagnosis. Alternate techniques have been used to determine the specific agent present in the histopathologic specimen, including immunohistochemistry, in situ hybridization, and PCR. In addition, techniques such as laser microdissection will be useful to detect the now more frequently recognized dual fungal infections and the local environment in which this phenomenon occurs.