Citations

Abstract Background Experts' interpretations of medical images for lesion diagnosis may not always align with the underlying in vivo tissue pathology and, therefore, cannot be considered the definitive truth regarding malignancy or benignity. While current machine learning (ML) models in medical imaging can replicate expert interpretations, their results may also diverge from the actual ground truth. Purpose This study investigates various factors contributing to these discrepancies and proposes solutions. Methods The central idea of the proposed solution is to integrate prior knowledge into ML models to enhance the characterization of in vivo tissues. The incorporation of prior knowledge into decision‐making is task‐specific, tailored to the data acquired for that task. This central idea was tested on the diagnosis of lesions using low dose computed tomography (LdCT) for early cancer detection, particularly focusing on more challenging, ambiguous or indeterminate lesions (IDLs) as classified by experts. One key piece of prior knowledge involves CT x‐ray energy spectrum, where different energies interact with in vivo tissues within a lesion, producing variable but reproducible image contrasts that encapsulate biological information. Typically, CT imaging devices use only the high‐energy portion of this spectrum for data acquisition; however, this study considers the full spectrum for lesion diagnostics. Another critical aspect of prior knowledge includes the functional or dynamic properties of in vivo tissues, such as elasticity, which can indicate pathological conditions. Instead of relying solely on abstract image features as current ML models do, this study extracts these tissue pathological characteristics from the image contrast variations. Results The method was tested on LdCT images of four sets of IDLs, including pulmonary nodules and colorectal polyps, with pathological reports serving as the ground truth for malignancy or benignity. The method achieved an area under the receiver operating characteristic curve (AUC) of 0.98 ± 0.03, demonstrating a significant improvement over existing state‐of‐the‐art ML models, which typically have AUCs in the 0.70 range. Conclusion Leveraging prior knowledge in machine intelligence can enhance lesion diagnosis, resolve the ambiguity of IDLs interpreted by experts, and improve the effectiveness of LdCT screening for early‐stage cancers.

Introduction: Endoscopic urethroplasty combines the minimally invasive benefit of endoscopic dilation or urethrotomy with the durability of open urethroplasty. Novel techniques in endoscopic urethroplasty have appeared through the years since the 1970s, including several techniques to hold free grafts against urethrotomy beds, such as grafts glued or sutured to a catheter, double-balloon catheter, and a novel balloon-graft carrying device along with transperineal sutures. 1 – 7 Ungerer et al. describe their endoscopic buccal graft urethroplasty using laparoscopic suturing instruments and a procedure-specific transurethral sheath. 8 In this video, we present our technique and early outcomes for transmeatal urethral surgery (TraMUS) for distal penile strictures. This technique applies to any stricture of the penile urethra where the distal aspect of a buccal mucosa graft (BMG) can be accessed transmeatally. Materials and Methods: The patient is prepped in the supine position. A 6 o’clock incision is made sharply using a nasal speculum with a scalpel, and an endoscopic cold knife set along the ventral plate of the urethra from the proximal end of the stricture to the meatus. Goal caliber is 26 French. The urethrotomy length is measured. A split-thickness BMG is deployed by passing a 26 French offset scope into the urethra. A Laborie needle is passed through the scope, through the mucosal edge of the urethrotomy, and out the penile skin. With each puncture, a 3-0 PDS free suture is passed through the needle, and the needle is removed to externalize the suture through the meatus. This is repeated until one pair of sutures exists on either side of the proximal urethrotomy. The equipment is removed and the transmeatal ends of each pair of sutures are passed through the proximal end of the graft from anti-mucosa to mucosa using a free needle and tied. The transcutaneous suture ends are pulled to parachute the graft into the urethra over the urethrotomy and then tied externally at the skin. The distal end of the graft is secured to the meatus with interrupted 4-0 Vicryl. Quilting sutures are placed along the graft using 3-0 Vicryl on skiied SH needles, similarly passing the needles through the urethra and out the skin with ends tied together, and then tying externally at the skin. Alternatively, quilting may be performed via the Laborie needle for more proximal placement. A Foley catheter is placed, and a compressive dressing is applied to the penis. The catheter remains for 2 weeks. Cystoscopy is performed at 4 months and 1 year to ensure patency. Results: Nineteen patients have been treated using TraMUS for distal penile strictures. Eighteen patients had at least 4 month of follow-up, with an average follow-up of 8 months (range 4–21 months). 2 failures are documented, in which we were unable to pass the 17 Fr flexible cystoscope. Median graft length was 5 cm (range 3–15 cm, IQR 4.9). Conclusions: TraMUS for distal penile strictures is a promising new technique for minimally invasive urethral reconstruction. The technique appears to be effective, and further data about outcomes are still needed. Patient Consent: Author(s) have received and archived patient consent for video recording/publication in advance of video recording of the procedure. Conflicts of Interest: The authors have no disclosures. Runtime of video: 8 mins 13 secs.

PURPOSE Sertoli-Leydig cell tumors (SLCTs) are rare sex cord-stromal tumors, representing <0.5% of all ovarian tumors. We analyze the role of germline DICER1 status in outcomes of ovarian SLCT. METHODS Patients with SLCT were enrolled in the International Pleuropulmonary Blastoma/ DICER1 Registry and/or the International Ovarian and Testicular Stromal Tumor Registry. Medical records were systematically abstracted, and those with known germline DICER1 status were selected for analysis. RESULTS Of 162 patients with SLCT, 60% had a germline DICER1 pathogenic or likely pathogenic (P/LP) variant. The adjusted 3-year recurrence-free survival (RFS) was 87.2% (95% CI, 79.4 to 95.8) for patients with a germline DICER1 P/LP variant compared with 78.1% (95% CI, 66.4 to 91.9) for those without a germline DICER1 P/LP variant ( P = .043). The adjusted 3-year and 5-year overall survival (OS) was 93.9% (95% CI, 87.3 to 100.0) for those with a germline DICER1 P/LP variant compared with the 3-year OS of 91.3% (95% CI, 83.4 to 100.0) and the 5-year OS of 78.2% (95% CI, 63.8 to 95.9) for those without a germline DICER1 P/LP variant ( P = .021). Among patients with a germline DICER1 P/LP variant, the risk of a subsequent, nonrecurrent event was 36.2% (95% CI, 21.4 to 48.1) within 10 years. Previous/concurrent and subsequent neoplasms were rare among those without a germline DICER1 P/LP variant. CONCLUSION This cohort study of patients with SLCT demonstrated that those with germline DICER1 P/LP variants had superior RFS and OS even when adjusting for other prognostic factors. Beyond prognostic implications of a germline DICER1 P/LP variant, germline testing helps identify patients at risk of subsequent neoplasms, including metachronous SLCT.

INTRODUCTION: The risk of progression to esophageal adenocarcinoma in Barrett's esophagus (BE) increases with advancing degrees of dysplasia. There is a critical need to improve the diagnosis of BE dysplasia, given substantial interobserver variability and overcalls of dysplasia during manual community pathologist reads. We aimed to externally validate a previously cross-validated BE dysplasia diagnosis deep learning model (BEDDLM) that predicts dysplasia grade on whole slide images (WSIs). METHODS: We digitized nondysplastic BE (NDBE), low-grade (LGD), and high-grade dysplasia (HGD) histology slides from 3 external academic centers. A consensus read by 2 expert study pathologists was used as the criterion standard. Slide stain characteristics were normalized using cycle-generative adversarial networks. WSIs were assessed by BEDDLM using an ensemble approach, combining a “You Only Look Once” model followed by a ResNet101 classifier model. RESULTS: We included 489 WSIs. Consensus histopathology revealed 232 NDBE, 117 LGD, and 140 HGD WSIs. The mean age (SD) was 66.9 (11.4) years; 413 (84.7%) were men. Using the BEDDLM ensemble model, sensitivity and specificity for NDBE were 73.3% (95% confidence interval [CI]: 67.09%–78.85%) and 93.4% (95% CI: 89.62%–96.10%); for LGD, 84.6% (95% CI: 76.78%–90.62%) and 80.6% (95% CI: 76.26%–84.54%); and for HGD, 80.7% (95% CI: 73.19%–86.89%) and 94.8% (95% CI: 91.97%–96.91%), respectively. The F1 score was 0.81, 0 0.69, and 0.83 for NDBE, LGD, and HGD, respectively. DISCUSSION: Our externally validated deep learning model demonstrates substantial accuracy for the diagnosis of BE dysplasia grade on WSIs.

Background Studies suggest that plasma Alzheimer's disease (AD) biomarkers may aid in the overall diagnosis of AD, but their utility among patients with atypical clinical presentations of AD are unknown. Objective The main objective of this study was to determine the relationship between amyloid-β (Aβ) and tau plasma biomarkers and PET measures of both Aβ and tau in atypical AD. The secondary objective was to determine if plasma biomarkers could differentiate patients with different atypical AD phenotypes and whether they were related to measures of disease severity. Methods We assessed whether plasma p-tau 181 and Aβ 42/40 were associated with Aβ and tau PET uptake, clinical phenotype and severity in 77 patients with PET biomarker-confirmed atypical AD. Results Plasma Aβ 42/40 ratio showed positive associations with tau PET uptake, with higher (more normal) Aβ 42/40 ratio associated with higher tau uptake; the ratio was not associated with Aβ PET. No associations were noted with plasma p-tau 181. Plasma Aβ 42/40 ratio and p-tau 181 concentrations were not associated with AD phenotype or cognitive severity. Conclusion Plasma Aβ 42/40 ratio and p-tau 181 concentrations are not associated with amyloid or tau PET or with clinical severity among individuals presenting with atypical AD.