CWCarl S. Winalski

Carl S. Winalski

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Carl S. Winalski has a total of 170 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Circumferential Meniscal Extrusion Mapping Using Radial Multiplanar Reconstructions

    by Ike R. Vayansky, Faysal Altahawi, Ahmet Hakan Ok, Richard Lartey, Sercan Tosun, Nancy Obuchowski, William Holden, Xiaojuan Li, Carl S. Winalski

    Meniscal extrusion is a marker of meniscal dysfunction and has been associated with osteoarthritis progression. Assessment is typically performed on a single coronal magnetic resonance imaging (MRI) slice, but this method may miss the location of maximal circumferential displacement. We developed an extrusion mapping technique using radially oriented multiplanar reconstruction (rMPR) images from high‐resolution 3D MRI to enable circumferential meniscal position measurements perpendicular to the tibial margin. Extrusion was measured by three independent readers on rMPR images generated from non‐fat‐saturated 3D fast spin‐echo MRI in 20 patients stratified by meniscal tear and extrusion status. Intrareader and interreader measurement variabilities were assessed using intraclass correlation coefficient (ICC), 95% repeatability coefficient (RC; intrareader), and 95% reproducibility coefficient (RDC; interreader). Corresponding rMPR and standard 2D coronal midbody extrusion measurements were compared. Intrareader variability was low, with a mean within‐meniscus within‐angle standard deviation (SD) of 0.2 mm and mean RC of 0.6 mm. Interreader variability was also low, with a mean within‐meniscus within‐angle SD of 0.5 mm and mean RDC of 1.4 mm. At the midbody, there was 98% concordance for extrusion classification between rMPR and 2D coronal measurements within the interreader RDC (1.4 mm). Exploratory regional analysis suggested different extrusion patterns between medial and lateral menisci and between torn and intact menisci. These findings suggest that circumferential meniscal extrusion mapping using rMPR images from 3D knee MRI is feasible, with low intrareader and interreader measurement variabilities. This framework enables regional characterization of meniscal displacement and should support future investigations of meniscus–osteoarthritis relationships. Statement of Clinical Significance: Meniscus extrusion circumferential mapping using radially oriented multiplanar reconstruction images is feasible and consistent across readers and allows a more complete assessment of meniscal position, which may be valuable for further investigation of meniscal injury and osteoarthritis.

  • Standardized Knee Meniscus MRI Reporting: An Interdisciplinary Delphi Consensus

    by Jie C. Nguyen, Vandan S. Patel, Geoffrey D. Abrams, Erin F. Alaia, Bethany U. Casagranda, Majid Chalian, Eric Y. Chang, Allison E. Crepeau, Kirkland W. Davis, Lutul D. Farrow, et al.

    Through a Delphi process, all 33 panelists reached consensus on topics relating to MRI reporting of knee meniscus lesions, demonstrating that interdisciplinary and multi-institutional standardization is possible.


  • Open Access

    Patellofemoral Cartilage Degradation Based on T1ρ Relaxation Times Varies Inversely With BMI After Patellar Dislocation

    by John J. Elias, Lutul D. Farrow, Richard Lartey, Ahmet Hakan Ok, Mei Li, Mingrui Yang, Carl S. Winalski, Xiaojuan Li

    Background: Elevated body mass index (BMI) is a primary factor contributing to osteoarthritis of the knee, but data are lacking on the relationship between BMI and cartilage properties shortly after a traumatic injury. Purpose/Hypothesis: The purpose of the study was to relate BMI to patellofemoral cartilage properties after a patellar dislocation, using magnetic resonance imaging–based T1ρ relaxation times to assess cartilage. It was hypothesized that high BMI is correlated with long patellofemoral cartilage T1ρ relaxation times, indicating cartilage degradation, in both control knees and knees after patellar dislocation. Study Design: Cross-sectional study; Level of evidence, 3. Methods: T1ρ relaxation time mapping was performed for underweight to obese (BMI < 40 kg/m 2 ) patients after a first-time patellar dislocation (n = 26; 19.8 ± 6.7 years old; 46 ± 36 days since dislocation), after multiple dislocations (n = 15; 18.4 ± 5.1 years old; 63 ± 33 days since most recent dislocation), and for healthy controls (n = 20; 20.1 ± 6.6 years old). For medial, lateral, and central regions of patellofemoral cartilage, mean T1ρ relaxation times were correlated against BMI. Statistical significance was set at P < .05. Results: For the control group, long T1ρ relaxation times were significantly correlated with high BMI for every region of patellofemoral cartilage except the lateral patella ( P < .025; r 2 = 0.26-0.63). On the contrary, for the first-time and multiple dislocations groups, long T1ρ relaxation times were significantly correlated with low BMI ( P < .05; r 2 = 0.24-0.50) for the medial patella (multiple dislocations only), central patella, lateral patella, medial trochlear groove (first-time dislocation only), central trochlear groove (first-time dislocation only), and lateral trochlear groove (first-time dislocation only). Conclusion: Long patellofemoral cartilage T1ρ relaxation times were related to high BMI for healthy knees, but unexpectedly related to low BMI after patellar dislocation. The results do not indicate a heightened risk of progressive cartilage degradation to osteoarthritis for overweight and obese patients immediately after patellar dislocation. For patients at a healthy weight and underweight, the traumatic injury is associated with early cartilage degradation, which could potentially initiate progressive degradation to posttraumatic osteoarthritis.


  • Open Access

    Unsupervised Segmentation of Knee Bone Marrow Edema-like Lesions Using Conditional Generative Models

    by Andrew Seohwan Yu, Mingrui Yang, Richard Lartey, William Holden, Ahmet Hakan Ok, Sameed Khan, Jeehun Kim, Carl Winalski, Naveen Subhas, Vipin Chaudhary, et al.

    Bone marrow edema-like lesions (BMEL) in the knee have been linked to the symptoms and progression of osteoarthritis (OA), a highly prevalent disease with profound public health implications. Manual and semi-automatic segmentations of BMELs in magnetic resonance images (MRI) have been used to quantify the significance of BMELs. However, their utilization is hampered by the labor-intensive and time-consuming nature of the process as well as by annotator bias, especially since BMELs exhibit various sizes and irregular shapes with diffuse signal that lead to poor intra- and inter-rater reliability. In this study, we propose a novel unsupervised method for fully automated segmentation of BMELs that leverages conditional diffusion models, multiple MRI sequences that have different contrast of BMELs, and anomaly detection that do not rely on costly and error-prone annotations. We also analyze BMEL segmentation annotations from multiple experts, reporting intra-/inter-rater variability and setting better benchmarks for BMEL segmentation performance.


  • Open Access

    Paper 31: Quantitative MRI of the Hamstring Muscles Ten Years After Autograft Hamstring ACLR

    by Richard Lartey, Nancy Obuchowski, Matthew Neill, Dongxing Xie, Brendan Eck, Mei Li, Jeehun Kim, Morgan Jones, Laura Huston, Kevin Harkins, et al.

    Objectives: After autograft hamstring harvesting for anterior cruciate ligament reconstruction (ACLR) there are several studies of semitendinosus (ST) and gracilis (G) tendon showing variation in muscle/tendon regeneration, muscle atrophy and myotendinous muscle retraction. There remains a gap in our knowledge on the extent of muscle function more than a decade after harvest. There is a paucity of long-term studies of hamstring muscle morphology following ACLR. There are no studies evaluating fatty infiltration, a known indicator of decreased muscle function. Likewise, the cross-sectional area of the hamstrings is a functional indicator of muscle strength. Finally, there are no comparative studies on hamstring and bone-patellar tendon-bone (BTB) autografts. In this study, we aim to determine the long-term function of the hamstring muscles as measured by quantitative magnetic resonance imaging (MRI) ten years after autograft hamstring ACLR by comparing to healthy controls and patients who received BTB autografts. Methods: The subjects were from the Multicenter Orthopaedic Outcome Network (MOON) nested cohort with at least ten-year follow-up and include 90 ACLR patients (50 hamstring autograft, 40 BTB autograft) and 21 healthy control participants. Inclusion criteria of the nested cohort were age 14-33 years old at index ACLR with unilateral ACL tear during a sport. Exclusion criteria were prior knee injury at enrollment or graft rupture during follow-up. Axial MR images of both legs from mid-thigh to knee were obtained for muscle cross-sectional area (CSA) and myotendinous junction position (MTJ) measurement. Six-point Dixon images for muscle %fat fraction measurement (%FF) were acquired at the same mid-thigh level as CSA measurements. The muscle CSA (cm2) of the semimembranosus (SM), semitendinosus (ST), gracilis (G), biceps femoris short head (Bshort), and biceps femoris long head (Blong) were measured on mid-thigh axial images by manual image segmentation. %FF maps were generated from the 6-point Dixon images and %FF of each hamstring muscle was calculated by overlying the muscle segmentation masks used for CSA measurement. ST and G MTJ distance (cm) to the joint line was measured manually using the axial and localizer images. For each graft type, a generalized linear model was first fit where the dependent variable was the biomarker measurement (separate models were fit for FF, CSA, and MTJ) and the independent variables were knee type (operated vs non-op), muscle, and their interaction. A transformation was used when the dependent variable was not normally distributed, as needed. Generalized Estimating Equations were used to account for the clustered nature of the data. A significance level of 0.05 was used to test the interaction term. If the interaction term was significant, then comparisons between operated and non-operated knees were performed for each muscle/tendon and graft type combination, using Holm’s method to control the type I error rate. Results: Patient demographics: age 33.5 ± 5.3 years; 45 female; body mass index [BMI] 25.8 ± 5.1 kg/m2; Controls demographics: age 30.3 ± 7.8 years; 14 female; BMI 24 ± 5.1 kg/m2. Mid-thigh muscle %FF (fat fraction (%)) values for hamstring autograft patients were significantly higher for the operated vs. non-operated limbs ST (15.2%±11.1% vs. 7.6%±5.2%) and G (13.1%±7.6% vs. 9.1%±4.4%) while the SM and biceps femoris showed no differences (Figure 1). In BTB autograft patients, there were no significant differences in muscle %FF between operated and non-operated limbs ST (7.3%±2.4% vs. 7.4%±2.5%) and G (10.1%±3.8% vs. 9.8%±3.9%) (Figure 1). For healthy controls overall, the left limbs had higher muscle %FF than the right limbs. Mid-thigh CSA (cross-sectional area (cm2)) measurements for hamstring autograft patients were significantly lower for the operated vs. non-operated limbs ST (6.7±3 vs 9.6±3.4) and G (3.7±1.8 vs. 4.5±1.6), while the SM and biceps femoris showed no differences (Figure 2). In BTB autograft patients, there were no significant differences in mid-thigh CSA between operated and non-operated limbs ST (10.2±3.7 vs. 10.4±4) and G (5.2±1.7 vs. 5.0±1.6) (Figure 2). For healthy controls, there were no significant differences in CSA between left vs. right limbs ST (7.9±2.6 vs. 7.9±2.4) and G (4.3±1.5 vs. 4.2±1.4). MTJ (myotendinous junction distance to the joint line (cm)) in hamstring autograft patients showed greater muscle retraction in the operated vs. non-operated limbs for ST (14.2±4.6 vs. 7.5±2.3) and G (9.9±4.4 vs. 6.4±1.6) (Figure 3). In BTB autograft patients, there were no significant differences in MTJ between operated vs. non-operated limbs ST (7.4±2.3 vs. 7.4±2) and G (6.2±1.5 vs. 6.2±1.4) (Figure 3). For controls there were also no significant differences in left vs. right limbs ST (5.7±2.5 vs. 5.6±2) and G (6.1±1.9 vs. 6.7±1.7). Conclusions: In long-term follow-up of over a decade following ACLR with autograft hamstring, the harvested hamstring muscles undergo significantly more fatty infiltration (higher muscle %FF), muscle atrophy (lower CSA) and myotendinous junction retraction (higher MTJ distance) than non-operated limbs in the same patients. Whereas the hamstring muscles remain symmetric as expected after ACLR with BTB autografts. No compensatory biceps femoris hypertrophy was found in hamstring autograft patients within this cohort. As a result of these three functional imaging measures (%FF, CSA, MTJ retraction) the harvested hamstrings do not resume normal indices. The biomechanical and functional consequences would need to be evaluated on an individual patient basis. One measure of harvest morbidity is the long-term persistence of significant changes in hamstring muscle functional imaging parameters after hamstring tendon harvesting for ACLR.

  • Open Access

    Poster 137: Influence of Body Mass Index on T1ρ Relaxation Times for Patellofemoral Cartilage: Patellar Dislocation Patients vs. Healthy Controls

    by Lutul Farrow, Mingrui Yang, Mei Li, Carl Winalski, Xiaojuan Li, John Elias

    Objectives: Patellar dislocations are associated with patellofemoral cartilage degradation following injury and a high risk of patellofemoral post-traumatic osteoarthritis (PTOA). Obesity is also a primary risk factor for patellofemoral cartilage degradation and OA, but the influence of obesity on cartilage degradation follow patellar dislocations is unknown. The current study uses T1ρ-based relaxation time mapping of patellofemoral cartilage to evaluate the influence of body mass index (BMI) on cartilage properties in the acute phase following a patellar dislocation, and compares the results to healthy knees. The hypothesis of the study is increasing BMI is associated with cartilage matrix degeneration for injured and healthy knees. Methods: The IRB-approved study focused on a BMI range from underweight to obese (< 40 kg/m2). The population included 18 patients being treated conservatively for a unilateral, first-time patellar dislocation (20 ± 8 years, 9 females, BMI: 24 ± 5 kg/m2, 56 ± 39 days since dislocation), 12 patients being treated conservatively for multiple unilateral dislocations (19 ± 6 years, 9 females, 24 ± 5 kg/m2, 67 ± 35 days since most recent dislocation), and 16 healthy controls (21 ± 7 years, 7 females; 21 ± 7 kg/m2). MRI scans included a 3D water excitation double echo steady state (DESS) scan of the knee for segmentation of the cartilage surfaces (slice thickness of 0.7 mm) and a T1ρ relaxation time scan (fat saturated scan, slice thickness = 4 mm, spin lock times = 0, 10, 40, and 80 ms, spin-lock frequency = 500 Hz). Long T1ρ relaxation times indicate low proteoglycan concentration within cartilage. Automated algorithms based on machine learning were used to segment cartilage on the patella and within the trochlear groove. Additional automated algorithms mapped T1ρ relaxation times to the reconstructed cartilage surfaces (Fig. 1). Mean T1ρ relaxation times were quantified for patellar and trochlear groove cartilage for each knee. Linear regression were used to correlate T1ρ relaxation times with BMI for the single dislocation, multiple dislocations, and control groups. Statistical significance was set at p < 0.05. Results: Patellofemoral T1ρ relaxation times increased with BMI for healthy control subjects, but decreased with BMI following patellar dislocation. For control subjects, the increase in T1ρ relaxation times with BMI was significant for cartilage on the patella and within the trochlear groove (r2 > 0.4, p < 0.01). For patients being treated for a single dislocation and multiple dislocations, the decrease in T1ρ relaxation times with increasing BMI was significant for cartilage on the patella and within the trochlear groove (r2 > 0.25, p < 0.025 for single dislocations, r2 > 0.3, p < 0.05 for multiple dislocations). Conclusions: The expected increase in T1ρ relaxation times with BMI was observed for healthy control subjects, but not for patients being treated for patellar dislocations. Elevated T1ρ values indicate loss of proteoglycan content that reduces cartilage integrity and could be an initiation point for progression to OA. For cartilage throughout the patellofemoral joint, T1ρ relaxation times decreased as BMI increased following patellar dislocation. The relationships were similar following one dislocation and multiple dislocations. The regression lines for injured knees and controls crossed at a BMI of approximately 25 for patellar and trochlear groove cartilage, indicating T1ρ relaxation times are higher for injured knees than healthy controls only for patients with a BMI < 25 (underweight or healthy weight). Multiple previous studies have identified elevated patellofemoral cartilage relaxation times from quantitative MRI following single and multiple patellar dislocations, but no study has previously attributed the difference to patients with a low BMI. The mechanism by which a high BMI limits cartilage degradation following patellar dislocation is currently unknown. The opposing relationships between T1ρ relaxation times and BMI for injured knees and controls occur despite similar ranges for T1ρ times and BMI. The primary difference between injured knees and controls is the recent dislocation event, and the ensuing inflammatory response. The single and multiple dislocation groups have a recent dislocation and inflammation in common despite differences in time since the first dislocation. Impact injury to cartilage is primarily limited to the medial patella, so is unlikely to cause the difference in trends. Pathologic patellofemoral anatomy is a characteristic of knees with patellar instability that could also influence the relationship with BMI. Additional enrollment and characterization of anatomical factors and inflammatory conditions is needed to identify factors that interact with BMI. Follow up analyses are also needed to determine if these relationships weaken with time from injury.

  • The Postoperative Knee: Cruciate and Other Ligaments

    by Yildiz Sengul, Kurt P. Spindler, Carl S. Winalski

  • Reproducibility of T1ρ and T2 quantification in a multi-vendor multi-site study

    by R. Lartey, A. Nanavati, J. Kim, M. Li, K. Xu, K. Nakamura, W. Shin, C.S. Winalski, N. Obuchowski, E. Bahroos, et al.


  • MRI Relaxometry as Early Measures of OA

    by Xiaojuan Li, Carl S. Winalski, Thomas M. Link