Shrinkage of Nasal Mucosa and Cartilage During Formalin Fixation

Deep Learning-Based Approach for Identifying and Measuring Focal Liver Lesions on Contrast-Enhanced MRI

BACKGROUND

The number of focal liver lesions (FLLs) detected by imaging has increased worldwide, highlighting the need to develop a robust, objective system for automatically detecting FLLs.

PURPOSE

To assess the performance of the deep learning-based artificial intelligence (AI) software in identifying and measuring lesions on contrast-enhanced magnetic resonance imaging (MRI) images in patients with FLLs.

STUDY TYPE

Retrospective.

SUBJECTS

395 patients with 1149 FLLs.

FIELD STRENGTH/SEQUENCE

The 1.5 T and 3 T scanners, including T1-, T2-, diffusion-weighted imaging, in/out-phase imaging, and dynamic contrast-enhanced imaging.

ASSESSMENT

The diagnostic performance of AI, radiologist, and their combination was compared. Using 20 mm as the cut-off value, the lesions were divided into two groups, and then divided into four subgroups: <10, 10-20, 20-40, and ≥40 mm, to evaluate the sensitivity of radiologists and AI in the detection of lesions of different sizes. We compared the pathologic sizes of 122 surgically resected lesions with measurements obtained using AI and those made by radiologists.

STATISTICAL TESTS

McNemar test, Bland-Altman analyses, Friedman test, Pearson's chi-squared test, Fisher's exact test, Dice coefficient, and intraclass correlation coefficients. A P-value <0.05 was considered statistically significant.

RESULTS

The average Dice coefficient of AI in segmentation of liver lesions was 0.62. The combination of AI and radiologist outperformed the radiologist alone, with a significantly higher detection rate (0.894 vs. 0.825) and sensitivity (0.883 vs. 0.806). The AI showed significantly sensitivity than radiologists in detecting all lesions <20 mm (0.848 vs. 0.788). Both AI and radiologists achieved excellent detection performance for lesions ≥20 mm (0.867 vs. 0.881, P = 0.671). A remarkable agreement existed in the average tumor sizes among the three measurements (P = 0.174).

DATA CONCLUSION

AI software based on deep learning exhibited practical value in automatically identifying and measuring liver lesions.

LEVEL OF EVIDENCE: 4

TECHNICAL EFFICACY

Stage 2.

Dependence of crack shape in loaded articular cartilage on the collagenous structure

Cartilage cracks disrupt tissue mechanics, alter cell mechanobiology, and often trigger tissue degeneration. Yet, some tissue cracks heal spontaneously. A primary factor determining the fate of tissue cracks is the compression-induced mechanics, specifically whether a crack opens or closes when loaded. Crack deformation is thought to be affected by tissue structure, which can be probed by quantitative polarized light microscopy (PLM). It is unclear how the PLM measures are related to deformed crack morphology. Here, we investigated the relationship between PLM-derived cartilage structure and mechanical behavior of tissue cracks by testing if PLM-derived structural measures correlated with crack morphology in mechanically indented cartilages.

METHODS

Knee joint cartilages harvested from mature and immature animals were used for their distinct collagenous fibrous structure and composition. The cartilages were cut through thickness, indented over the cracked region, and processed histologically. Sample-specific birefringence was quantified as two-dimensional (2D) maps of azimuth and retardance, two measures related to local orientation and degree of alignment of the collagen fibers, respectively. The shape of mechanically indented tissue cracks, measured as depth-dependent crack opening, were compared with azimuth, retardance, or "PLM index," a new parameter derived by combining azimuth and retardance.

RESULTS

Of the three parameters, only the PLM index consistently correlated with the crack shape in immature and mature tissues.

CONCLUSION

In conclusion, we identified the relative roles of azimuth and retardance on the deformation of tissue cracks, with azimuth playing the dominant role. The applicability of the PLM index should be tested in future studies using naturally-occurring tissue cracks.

Development of an Animal Model for Type II Sulcus (Sulcus Vergeture)

OBJECTIVES

To develop a reproducible experimental animal model for sulcus vergeture in rabbits.

STUDY DESIGN

Experimental animal study.

MATERIAL METHOD

We evaluated three methods of inducing sulcus in twelve New Zealand white rabbits to produce a sulcus model. Experimental groups comprised: group 1 (n = 4) underwent submucosal injury following endolaryngeal epithelial incision; group 2 (n = 4) received submucosal injury via thyrotomy; group 3 (n = 4) received submucosal injury via thyrotomy, followed with horizontal mucosal lateralization suture. Right vocal folds of the rabbits underwent surgery to produce sulcus vocalis and left vocal folds were used for the comparison. In the sixth week after the procedure, the rabbits were sacrificed and the larynxes were harvested and analyzed histopathologically.

RESULTS

No animals in group 1 or 2 developed sulcus vocalis. Sulcus formation was observed in all rabbits in group 3, under endoscopic examination and microscopic sulcus formation was demonstrated for the first, second, and third rabbits. An epithelial depression area was seen at the glandular ductal opening zone for the fourth rabbit, but it could also be accepted as a sulcus formation.

CONCLUSION

We successfully developed a reproducible survival model for sulcus using a submucosal injury via thyrotomy, followed with a horizontal mucosal lateralization suture. This model provides the groundwork for future research into the applicability of new approaches for sulcus management.

Deformation behaviors and mechanical impairments of tissue cracks in immature and mature cartilages

Degeneration of articular cartilage is often triggered by a small tissue crack. As cartilage structure and composition change with age, the mechanics of cracked cartilage may depend on the tissue age, but this relationship is poorly understood. Here, we investigated cartilage mechanics and crack deformation in immature and mature cartilage exposed to a full-thickness tissue crack using indentation testing and histology, respectively. When a cut was introduced, tissue cracks opened wider in the mature cartilage compared to the immature cartilage. However, the opposite occurred upon mechanical indentation over the cracked region. Functionally, the immature-cracked cartilages stress-relaxed faster, experienced increased tissue strain, and had reduced instantaneous stiffness, compared to the mature-cracked cartilages. Taken together, mature cartilage appears to withstand surface cracks and maintains its mechanical properties better than immature cartilage and these superior properties can be explained by the structure of their collagen fibrous network.

Lateral Meniscus Height and ACL Reconstruction Failure: A Nested Case-Control Study

Previous work has shown that the morphology of the knee joint is associated with the risk of primary anterior cruciate ligament (ACL) injury. The objective of this study is to analyze the effect of the meniscal height, anteroposterior distance of the lateral tibial plateau, and other morphological features of the knee joint on risk of ACL reconstruction failure. A nested case-control study was conducted on patients who underwent an ACL reconstruction surgery during the period between 2008 and 2015. Cases were individuals who failed surgery during the study period. Controls were patients who underwent primary ACL reconstruction surgery successfully during the study period. They were matched by age (±2 years), gender, surgeon, and follow-up time (±1 year). A morphological analysis of the knees was then performed using the preoperative magnetic resonance imaging scans. The anteroposterior distance of the medial and lateral tibial plateaus was measured on the T2 axial cuts. The nonweightbearing maximum height of the posterior horn of both menisci was measured on the T1 sagittal scans. Measurements of the medial and lateral tibial slope and meniscal slope were then taken from the sagittal T1 scans passing through the center of the medial and lateral tibial plateau. A binary logistic regression analysis was done to calculate crude and adjusted odds ratios (ORs) estimates. Thirty-four cases who underwent ACL revision surgery were selected and were matched with 68 controls. Cases had a lower lateral meniscal height (6.39 ± 1.2 vs. 7.02 ± 0.9, p = 0.008, power = 84.4%). No differences were found between the two groups regarding the bone slope of the lateral compartment (6.19 ± 4.8 vs. 6.92 ± 5.8, p = 0.552), the lateral meniscal slope (-0.28 ± 5.8 vs. -1.03 ± 4.7, p = 0.509), and the anteroposterior distance of the lateral tibial plateau (37.1 ± 5.4 vs. 35.6 ± 4, p = 0.165). In addition, no differences were found in the medial meniscus height between cases and controls (5.58 ± 1.2 vs. 5.81 ± 1.2, respectively, p = 0.394). There were also no differences between cases and controls involving the medial bone slope, medial meniscal slope, or anterior posterior distance of the medial tibial plateau. Female patients had a higher medial (4.8 degrees ± 3.2 vs. 3.3 ± 4.1, p = 0.047) and lateral (8.1 degrees ± 5.1 vs. 5.6 degrees  ± 5.6, p = 0.031) tibial bone slope, and a lower medial (5.3 mm ± 1.0 vs. 6.1 mm ± 1.2, p = 0.001) and lateral (6.6 ± 1.0 vs. 7.0 ± 1.2, p = 0.035) meniscus height, and medial (4.3 ± 0.4 vs. 4.8 ± 0.4, p =0.000) and lateral (3.3 ± 0.3 vs. 3.9 ± 0.4, p = 0.000) anteroposterior distance than males, respectively.The adjusted OR of suffering an ACL reconstruction failure compared to controls was 5.1 (95% confidence interval [CI]: 1.7-14.9, p = 0.003) for patients who had a lateral meniscus height under 6.0 mm. The adjusted OR of suffering an ACL reconstruction failure was 2.4 (95% CI: 1.0-7.7, p = 0.01) for patients who had an anteroposterior distance above 35.0 mm. Patients with a lateral meniscal height under 6.0 mm have a 5.1-fold risk of suffering an ACL reconstruction failure compared to individuals who have a lateral meniscal height above 6.0 mm. Patients with a higher anteroposterior distance of the lateral tibial plateau also have a higher risk of ACL reconstruction failure.

Pancreatic adenocarcinoma: variability in measurements of tumor size among computed tomography, magnetic resonance imaging, and pathologic specimens

PURPOSE

To compare the tumor size measurements assessed by computed tomography (CT) and magnetic resonance imaging (MRI) versus measurements of resected pathologic specimens from patients with pancreatic ductal adenocarcinoma (PDAC).

METHODS

This study included 114 patients with histologically confirmed PDAC who underwent contrast-enhanced CT and MRI before surgery. The tumor sizes from CT, MRI, and pathologic specimens were compared by using Bland-Altman analyses and intraclass correlation coefficients (ICCs). The discrepancies in PDAC size between CT/MRI and pathologic specimens were calculated and contributing factors for the discrepancies, including tumor locations (pancreatic head/neck, body, or tail), T stages (T1, T2, or T3), and N stages (N0, N1, or N2), were analyzed with Pearson's correlation coefficients and multivariable linear regression analyses.

RESULTS

There was significant difference among the mean tumor sizes of three measurements (P < 0.001). The difference in mean tumor size between the pathologic sizes for PDAC was 4.3 mm (ICC 0.67) on CT and 5.8 mm (ICC 0.65) on MRI. Both CT and MRI showed wide ranges of limits of agreement (LOAs) between the pathologic specimens for tumor size measurements (LOAs, - 28.9 to 21.4 and - 29.4 to 17.8, respectively). The tumor size on CT or MRI was estimated to be smaller than that on pathology when the tumor was > 30 mm. The discrepancies in the tumor size estimated between CT/MRI and pathologic specimens were significantly different for tumors of different T stages (P < 0.001).

CONCLUSIONS

Both contrast-enhanced CT and MRI underestimate the mean tumor size by 4.3 mm and 5.8 mm, respectively, compared to the size of pathologic specimens. A larger tumor size indicates a greater discrepancy in the PDAC size measurements between CT/MRI and pathologic specimens.

Formulation and Evaluation of Eudragit RL-100 Nanoparticles Loaded In-Situ Forming Gel for Intranasal Delivery of Rivastigmine

Purpose: Rivastigmine hydrogen tartrate (RHT) is commonly used for the treatment of mild to moderate Alzheimer’s disease (AD). The aim of this work was to formulate in-situ pluronic F-127 (PF-127) hydrogels containing Eudragit RL-100 (EU-RL) nanoparticles (NPs) in order to improve the therapeutic efficacy of RHT through the nasal route.

Methods: The NPs were prepared using different polymer to drug ratios and evaluated for their physicochemical characteristics, cellular uptake and in vitro cytotoxicity against lung adenocarcinoma cells (A459). PF-127 nanoformulations were prepared via cold method and analyzed in terms of physicochemical properties and drug release profiles. The nanoformulations and plain drug gel were then assessed by ex vivo permeation studies across the sheep nasal mucosa.

Results: The EU-RL NPs exhibited a particle size within the range of 118 to 154 nm and positive zeta potential values of 22.5 to 30 mV with an approximately spherical shape. Fourier transform infrared spectroscopy (FTIR), differential scanning calorimetry (DSC), and X-ray powder diffraction (XRPD) suggested no drug to polymer interaction through the preparation of nanoformulations. The RHT-loaded NPs exhibited an acceptable cytocompatibility with a time- and dose-dependent cellular internalization.

Conclusion: Our results clearly indicated the potential of nanoformulations as controlled release systems to improve the therapeutic efficacy of RHT through the intranasal administration

The evolving large-strain shear responses of progressively osteoarthritic human cartilage

OBJECTIVE

The composition and structure of articular cartilage evolves during the development and progression of osteoarthritis (OA) resulting in changing mechanical responses. We aimed to assess the evolution of the intrinsic, large-strain mechanics of human articular cartilage-governed by collagen and proteoglycan and their interactions-during the progression of OA.

DESIGN

We completed quasi-static, large-strain shear tests on 64 specimens from ten donors undergoing total knee arthroplasty (TKA), and quantified the corresponding state of OA (OARSI grade), structural integrity (PLM score), and composition (glycosaminoglycan and collagen content).

RESULTS

We observed nonlinear stress-strain relationships with distinct hystereses for all magnitudes of applied strain where stiffnesses, nonlinearities, and hystereses all reduced as OA advanced. We found a reduction in energy dissipation density up to 80% in severely degenerated (OARSI grade 4, OA-4) vs normal (OA-1) cartilage, and more importantly, we found that even cartilage with a normal appearance in structure and composition (OA-1) dissipated 50% less energy than healthy (control) load-bearing cartilage (HL₀). Changes in stresses and stiffnesses were in general less pronounced and did not allow us to distinguish between healthy load-bearing controls and very early-stage OA (OA-1), or to distinguish consistently among different levels of degeneration, i.e., OARSI grades.

CONCLUSIONS

Our results suggest that reductions in energy dissipation density can be detected by bulk-tissue testing, and that these reductions precede visible signs of degeneration. We highlight the potential of energy dissipation, as opposed to stress- or stiffness-based measures, as a marker to diagnose early-stage OA.

Longitudinal study of rat volar fat pad fixation and ethanol storage: implications for the use of fluid-preserved specimens in morphological studies

Museum fluid collections preserve important biological specimens for study. Tissues are often fixed in 10% buffered formalin to halt metabolic activities and transferred to a solution of ethanol for long-term storage. This process, however, forces water from the tissues and has been shown to alter the morphology of preserved specimens in ways that may influence the biological interpretation of results. The degree to which fluid preservation alters morphology is linked to multiple biological factors, such as tissue size and composition, and should therefore be examined prior to functional analysis. This study is undertaken as part of a more inclusive examination of mammalian volar morphology. A sample of five adult male and five adult female rats (Rattus norvegicus) was utilized to evaluate longitudinal changes in the dimensions of the volar pads across fixation in 10% buffered formalin and preservation in 70% ethanol for 1 year. No significant changes to the measured dimensions of the rat volar pads were present across stages of fixation and preservation, and no significant interactions of specimen size or sex were noted. These findings indicate that small mammalian volar pads that have been fixed in 10% buffered formalin and stored in 70% ethanol are appropriate for morphological study using the measurements described here without corrective algorithms. This finding is rare among preservation studies but highlights the variability of tissue behavior during chemical preservation and the necessity of preliminary investigations of preservation artifacts. Concurrence here between the preserved and unpreserved samples is likely related to the anhydrous nature of the volar pads and the supporting skeletal structure, and their confined position between major joints of the hands and feet.

Effect of Measurement Technique on TMJ Mandibular Condyle and Articular Disc Morphometry: CBCT, MRI, and Physical Measurements

PURPOSE

Accurate description of the temporomandibular size and shape (morphometry) is critical for clinical diagnosis and surgical planning and the design and development of regenerative scaffolds and prosthetic devices and to model the temporomandibular loading environment. The study objective was to determine the 3-dimensional morphometry of the temporomandibular joint (TMJ) condyle and articular disc using cone-beam computed tomography (CBCT), magnetic resonance imaging (MRI), and physical measurements of the same joints using a repeated measures design and to determine the effect of the measurement technique on temporomandibular size and shape.

MATERIALS AND METHODS

Human cadaveric heads underwent a multistep protocol to acquire physiologically meaningful measurements of the condyle and disc. The heads first underwent CBCT scanning, and solid models were automatically generated. The superficial soft tissues were dissected, and intact TMJs were excised and underwent MRI scanning, with solid models generated after manual segmentation. After MRI, the intact joints were dissected, and physical measurements of the condyle and articular disc were performed. The CBCT-based model measurements, MRI-based model measurements, and physical measurements were standardized, and a repeated measures study design was used to determine the effect of the measurement technique on the morphometric parameters.

RESULTS

Multivariate general linear mixed effects models showed significant effects for measurement technique for condylar morphometric outcomes (P < .001) and articular disc morphometric outcomes (P < .001). The physical measurements after dissection were larger than either the CBCT-based or MRI-based measurements. Differences in imaging-based morphometric parameters followed a complex relationship between imaging modality resolution and contrast between tissue types.

CONCLUSIONS

Physical measurements after dissection are still considered the reference standard. However, owing to their inaccessibility in vivo, understanding how the imaging technique affects the temporomandibular size and shape is critical toward the development of high-fidelity solid models to be used in the design and development of regenerative scaffolds, surgical planning, prosthetic devices, and anatomic investigations.