Real-World Financial and Clinical Impact of Diagnostic-Driven and Empirical-Treatment Strategies in High-Risk Immunocompromised Patients with Suspected Aspergillus Infection in the United Kingdom

Machine learning to identify clinically relevant Candida yeast species

Fungal infections, especially due to Candida species, are on the rise. Multi-drug resistant organisms such as Candida auris are difficult and time consuming to identify accurately. Machine learning is increasingly being used in health care, especially in medical imaging. In this study, we evaluated the effectiveness of six convolutional neural networks (CNNs) to identify four clinically important Candida species. Wet-mounted images were captured using bright field live-cell microscopy followed by separating single-cells, budding-cells, and cell-group images which were then subjected to different machine learning algorithms (custom CNN, VGG16, ResNet50, InceptionV3, EfficientNetB0, and EfficientNetB7) to learn and predict Candida species. Among the six algorithms tested, the InceptionV3 model performed best in predicting Candida species from microscopy images. All models performed poorly on raw images obtained directly from the microscope. The performance of all models increased when trained on single and budding cell images. The InceptionV3 model identified budding cells of C. albicans, C. auris, C. glabrata (Nakaseomyces glabrata), and C. haemulonii in 97.0%, 74.0%, 68.0%, and 66.0% cases, respectively. For single cells of C. albicans, C. auris, C. glabrata, and C. haemulonii InceptionV3 identified 97.0%, 73.0%, 69.0%, and 73.0% cases, respectively. The sensitivity and specificity of InceptionV3 were 77.1% and 92.4%, respectively. Overall, this study provides proof of the concept that microscopy images from wet-mounted slides can be used to identify Candida yeast species using machine learning quickly and accurately.

Comparing the Real-World Use of Isavuconazole to Other Anti-Fungal Therapy for Invasive Fungal Infections in Patients with and without Underlying Disparities: A Multi-Center Retrospective Study

Invasive fungal infections (IFIs) are a major cause of morbidity and mortality among immunocompromised patients with underlying malignancies and prior transplants. FDA approved Isavuconazole as a primary therapy for Invasive Aspergillosis (IA) and Mucormycosis. This study aims to compare the real-world clinical outcomes and safety of isavuconazole to voriconazole and an amphotericin B-based regimen in patients with underlying malignancies and a transplant. In addition, the response to anti-fungal therapy and the outcome were compared among patients with a disparity (elderly, obese patients, patients with renal insufficiency and diabetes mellitus) versus those with no disparity. We performed a multicenter retrospective study, including patients with cancer diagnosed with an invasive fungal infection, and treated primarily with isavuconazole, voriconazole or amphotericin B. Clinical, radiologic findings, response to therapy and therapy related adverse events were evaluated during 12 weeks of follow-up. We included 112 patients aged 14 to 77 years, and most of the IFIs were classified into definite (29) or probable (51). Most cases were invasive aspergillosis (79%), followed by fusariosis (8%). Amphotericin B were used more frequently as primary therapy (38%) than isavuconazole (30%) or voriconazole (31%). Twenty one percent of the patients presented adverse events related to primary therapy, with patients receiving isavuconazole presenting less adverse events when compared to voriconazole and amphotericin (p < 0.001; p = 0.019). Favorable response to primary therapy during 12 weeks of follow-up were similar when comparing amphotericin B, isavuconazole or voriconazole use. By univariate analysis, the overall cause of mortality at 12 weeks was higher in patients receiving amphotericin B as primary therapy. However, by multivariate analysis, Fusarium infection, invasive pulmonary infection or sinus infection were the only independent risk factors associated with mortality. In the treatment of IFI for patients with underlying malignancy or a transplant, Isavuconazole was associated with the best safety profile compared to voriconazole or amphotericin B-based regimen. Regardless of the type of anti-fungal therapy used, invasive Fusarium infections and invasive pulmonary or sinus infections were the only risk factors associated with poor outcomes. Disparity criteria did not affect the response to anti-fungal therapy and overall outcome, including mortality.