A systematic review on the use of artificial intelligence in gynecologic imaging - Background, state of the art, and future directions.

OBJECTIVE: Machine learning, deep learning, and artificial intelligence (AI) are terms that have made their way into nearly all areas of medicine. In the case of medical imaging, these methods have become the state of the art in nearly all areas from image reconstruction to image processing and automated analysis. In contrast to other areas, such as brain and breast imaging, the impacts of AI have not been as strongly felt in gynecologic imaging. In this review article, we: (i) provide a background of clinically relevant AI concepts, (ii) describe methods and approaches in computer vision, and (iii) highlight prior work related to image classification tasks utilizing AI approaches in gynecologic imaging. DATA SOURCES: A comprehensive search of several databases from each database's inception to March 18th, 2021, English language, was conducted. The databases included Ovid MEDLINE(R) and Epub Ahead of Print, In-Process & Other Non-Indexed Citations, and Daily, Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, and Ovid Cochrane Database of Systematic Reviews and ClinicalTrials.gov. METHODS OF STUDY SELECTION: We performed an extensive literature review with 61 articles curated by three reviewers and subsequent sorting by specialists using specific inclusion and exclusion criteria. TABULATION, INTEGRATION, AND RESULTS: We summarize the literature grouped by each of the three most common gynecologic malignancies: endometrial, cervical, and ovarian. For each, a brief introduction encapsulating the AI methods, imaging modalities, and clinical parameters in the selected articles is presented. We conclude with a discussion of current developments, trends and limitations, and suggest directions for future study. CONCLUSION: This review article should prove useful for collaborative teams performing research studies targeted at the incorporation of radiological imaging and AI methods into gynecological clinical practice.

Synergy Between Pseudomonas aeruginosa Filtrates And Voriconazole Against Aspergillus fumigatus Biofilm Is Less for Mucoid Isolates From Persons With Cystic Fibrosis.

Persons with cystic fibrosis (CF) frequently suffer from Pseudomonas aeruginosa and Aspergillus fumigatus co-infections. There is evidence that co-infections with these interacting pathogens cause airway inflammation and aggravate deterioration of lung function. We recently showed that P. aeruginosa laboratory isolates synergistically interact with the anti-fungal azole voriconazole (VCZ), inhibiting biofilm metabolism of several A. fumigatus laboratory strains. Interaction was usually mediated via pyoverdine, but also via pyocyanin or pyochelin. Here we used planktonic filtrates of 7 mucoid and 9 non-mucoid P. aeruginosa isolates from CF patients, as well as 8 isolates without CF origin, and found that all of these isolates interacted with VCZ synergistically at their IC50 as well as higher dilutions. CF mucoid isolates showed the weakest interactive effects. Four non-mucoid P. aeruginosa CF isolates produced no or very low levels of pyoverdine and did not reach an IC50 against forming A. fumigatus biofilm; interaction with VCZ still was synergistic. A VCZ-resistant A. fumigatus strain showed the same level of susceptibility for P. aeruginosa anti-fungal activity as a VCZ-susceptible reference strain. Filtrates of most Pseudomonas isolates were able to increase anti-fungal activity of VCZ on a susceptible A. fumigatus strain. This was also possible for the VCZ-resistant strain. In summary these data show that clinical P. aeruginosa isolates, at varying degrees, synergistically interact with VCZ, and that pyoverdine is not the only molecule responsible. These data also strengthen the idea that during co-infections of A. fumigatus and P. aeruginosa lower concentrations of VCZ might be sufficient to control fungal growth.

Altered Pseudomonas Strategies to Inhibit Surface Aspergillus Colonies.

Pseudomonas aeruginosa and Aspergillus fumigatus infections frequently co-localize in lungs of immunocompromised patients and individuals with cystic fibrosis (CF). The antifungal activity of P. aeruginosa has been described for its filtrates. Pyoverdine and pyocyanin are the principal antifungal P. aeruginosa molecules active against A. fumigatus biofilm metabolism present in iron-limited or iron-replete planktonic P. aeruginosa culture filtrates, respectively. Using various P. aeruginosa laboratory wild-type strains (PA14, PAO1, PAK), we found antifungal activity against Aspergillus colonies on agar. Comparing 36 PA14 and 7 PAO1 mutants, we found that mutants lacking both major siderophores, pyoverdine and pyochelin, display higher antifungal activity on agar than their wild types, while quorum sensing mutants lost antifungal activity. Addition of ferric iron, but not calcium or magnesium, reduced the antifungal effects of P. aeruginosa on agar, whereas iron-poor agar enhanced antifungal effects. Antifungal activity on agar was mediated by PQS and HHQ, via MvfR. Among the MvfR downstream factors, rhamnolipids and elastase were produced in larger quantities by pyoverdine-pyochelin double mutants and showed antifungal activity on agar. In summary, antifungal factors produced by P. aeruginosa on agar differ from those produced by bacteria grown in liquid cultures, are dependent on quorum sensing, and are downregulated by the availability of ferric iron. Rhamnolipids and elastase seem to be major mediators of Pseudomonas' antifungal activity on a solid surface.

Efficacy of nikkomycin Z in murine CNS coccidioidomycosis: modelling sustained-release dosing.

OBJECTIVES: Meningitis is the most feared coccidioidomycosis complication. Nikkomycin Z (nikZ) is a chitin synthase inhibitor. A concern is short half-life, necessitating multiple dose/day regimens. We simulated extended release, providing nikZ in drinking water. Extended release would enhance convenience, and adherence, for patients. METHODS: Coccidioides posadasii was injected intracerebrally into mice. Twelve day treatments began on Day 3. Fluconazole was given 100 mg/kg once daily (gavage); designed doses of nikZ 30, 100 or 300 mg/kg/day in drinking water. On Day 30 post-treatment, survivors were euthanized, brain cfu quantitated and cfu in other organs assessed. RESULTS: nikZ was stable in drinking water. Survival was 11%, 50%, 70%, 90% and 100% in untreated controls, fluconazole and nikZ 30, 100 and 300 mg/kg/day, respectively ; nikZ 300 mg/kg/day was superior (P ≤ 0.01) to fluconazole. Brains were sterilized in 0%, 20%, 86%, 89% and 80% of mice, respectively; nikZ 100 or 300 mg/kg/day was superior (P ≤ 0.01) to fluconazole. Clearance of infection in other organs was similar. All decreased drinking after infection, causing nikZ mice to ingest less than the desired dose in early therapy; despite this, they recovered sufficiently to resume pre-infection drinking and designed drug intakes. Thus, when sickest, even less than the designed dose was sufficient to enable recovery. CONCLUSIONS: This efficacy supports the development of sustained-release nikZ. Decreased intake wouldn't be a factor in humans, receiving drug via extended-release pill or continuous IV infusion. In prior studies (twice daily nikZ) of murine coccidioidal meningitis, results were inferior, suggesting sustained release may provide both convenience and superior outcomes.

Pseudomonas aeruginosa Virulence Factors Support Voriconazole Effects on Aspergillus fumigatus.

Pseudomonas aeruginosa and Aspergillus fumigatus are pathogens that are associated with deterioration of lung function, e.g., in persons with cystic fibrosis (CF). There is evidence that co-infections with these pathogens cause airway inflammation and aggravate pathology in CF lungs. Intermicrobial competition of P. aeruginosa and A. fumigatus has been described, but it is unknown how anti-fungal therapy is affected. The anti-fungal azole voriconazole (VCZ), supernatants of P. aeruginosa laboratory isolates PA14 or PAO1, or clinical isolate Pa10 independently inhibited biofilm metabolism of A. fumigatus isolates 10AF and AF13073. When VCZ and supernatants were combined at their IC50s, synergistic effects on A. fumigatus were found. Synergistic effects were no longer observed when P. aeruginosa supernatants were prepared in the presence of iron, or when P. aeruginosa mutants were lacking the ability to produce pyoverdine and pyochelin. Combination of pure P. aeruginosa products pyoverdine, pyochelin, and pyocyanin with VCZ showed synergistic anti-fungal effects. Combining VCZ with P. aeruginosa supernatants also improved its MIC and MFC against planktonic A. fumigatus. In summary, in the case of P. aeruginosa-A. fumigatus co-infections, it appeared that the P. aeruginosa co-infection facilitated therapy of the Aspergillus; lower concentrations of VCZ might be sufficient to control fungal growth.