Machine Learning Detection and Characterization of Splenic Injuries on Abdominal Computed Tomography.

BACKGROUND: Multi-detector contrast-enhanced abdominal computed tomography (CT) allows for the accurate detection and classification of traumatic splenic injuries, leading to improved patient management. Their effective use requires rapid study interpretation, which can be a challenge on busy emergency radiology services. A machine learning system has the potential to automate the process, potentially leading to a faster clinical response. This study aimed to create such a system. METHOD: Using the American Association for the Surgery of Trauma (AAST), spleen injuries were classified into 3 classes: normal, low-grade (AAST grade I-III) injuries, and high-grade (AAST grade IV and V) injuries. Employing a 2-stage machine learning strategy, spleens were initially segmented from input CT images and subsequently underwent classification via a 3D dense convolutional neural network (DenseNet). RESULTS: This single-centre retrospective study involved trauma protocol CT scans performed between January 1, 2005, and July 31, 2021, totaling 608 scans with splenic injuries and 608 without. Five board-certified fellowship-trained abdominal radiologists utilizing the AAST injury scoring scale established ground truth labels. The model achieved AUC values of 0.84, 0.69, and 0.90 for normal, low-grade injuries, and high-grade splenic injuries, respectively. CONCLUSIONS: Our findings demonstrate the feasibility of automating spleen injury detection using our method with potential applications in improving patient care through radiologist worklist prioritization and injury stratification. Future endeavours should concentrate on further enhancing and optimizing our approach and testing its use in a real-world clinical environment.

Interrater Agreement of CT Grading of Blunt Splenic Injuries: Does the AAST Grading Need to Be Reimagined?

Introduction: The Revised Organ Injury Scale (OIS) of the American Association for Surgery of Trauma (AAST) is the most widely accepted classification of splenic trauma. The objective of this study was to evaluate inter-rater agreement for CT grading of blunt splenic injuries. Methods: CT scans in adult patients with splenic injuries at a level 1 trauma centre were independently graded by 5 fellowship trained abdominal radiologists using the AAST OIS for splenic injuries - 2018 revision. The inter-rater agreement for AAST CT injury score, as well as low-grade (IIII) versus high-grade (IV-V) splenic injury was assessed. Disagreement in two key clinical scenarios (no injury versus injury, and high versus low grade) were qualitatively reviewed to identify possible sources of disagreement. Results: A total of 610 examinations were included. The inter-rater absolute agreement was low (Fleiss kappa statistic 0.38, P < 0.001), but improved when comparing agreement between low and high grade injuries (Fleiss kappa statistic of 0.77, P < .001). There were 34 cases (5.6%) of minimum two-rater disagreement about no injury vs injury (AAST grade ≥ I). There were 46 cases (7.5%) of minimum two-rater disagreement of low grade (AAST grade I-III) versus high grade (AAST grade IV-V) injuries. Likely sources of disagreement were interpretation of clefts versus lacerations, peri-splenic fluid versus subcapsular hematoma, application of adding multiple low grade injuries to higher grade injuries, and identification of subtle vascular injuries. Conclusion: There is low absolute agreement in grading of splenic injuries using the existing AAST OIS for splenic injuries.

Delayed Gadolinium Leakage in Ocular Structures on Brain MR Imaging: Prevalence and Associated Factors.

BACKGROUND AND PURPOSE: Gadolinium leakage in ocular structures (GLOS) is characterized by hyperintense signal in the chambers of the eye on FLAIR and has been reported in association with blood-ocular barrier breakdown in patients with ischemic strokes. The underlying mechanism of GLOS remains poorly understood; however, some studies suggest it may be part of a physiologic excretion pathway of gadolinium. This study aimed to determine the prevalence of GLOS in an unselected patient population. MATERIALS AND METHODS: A retrospective analysis was conducted on 439 patients who underwent brain MR imaging within 7 days of receiving a gadolinium-based contrast agent injection for a prior MR imaging study. Clinical, imaging, and laboratory data were collected. Descriptive and logistic regression analyses were performed. RESULTS: GLOS was observed in 26 of 439 patients (6%). The occurrence of GLOS varied with time, with 3 (12%), 14 (54%), 8 (31%), and 1 (4%) patient showing GLOS within 24, 25-72, 73-120, and >120 hours after gadolinium-based contrast agent injection, respectively. Patients with GLOS were older (median age: 72 versus 55 years, P = .001) and had higher median serum creatinine levels (73 versus 64 µmol/L, P = .005) and a lower median estimated glomerular filtration rate (84 versus 101 mL/min/1.73 m2, P < .001). A shorter median time interval between gadolinium-based contrast agent injection and the index brain MR imaging was observed in the group positive for GLOS (62 versus 91 hours, P = .003). Multivariable regression analysis identified the estimated glomerular filtration rate (OR = 0.970; 95% CI, 0.049-0.992; P = .008) and time interval since gadolinium-based contrast agent injection (OR = 0.987; 95% CI, 0.977-0.997; P = .012) as independent factors associated with GLOS. CONCLUSIONS: GLOS was observed in only a small percentage of patients receiving gadolinium-based contrast agent within 7 days before brain MR imaging. This phenomenon was noted in patients with normal findings on brain MR imaging and those with various CNS pathologies, and it was associated with lower estimated glomerular filtration rates and shorter time intervals after gadolinium-based contrast agent injection. While GLOS may be a physiologic gadolinium-based contrast agent excretion pathway, the presence of ocular disease was not formally evaluated in the included population. Awareness of GLOS is nonetheless useful for appropriate radiologic interpretation.

Evaluation of quantitative MRCP (MRCP+) for risk stratification of primary sclerosing cholangitis: comparison with morphological MRCP, MR elastography, and biochemical risk scores.

OBJECTIVES: To study the association of MRCP+ parameters with biochemical scoring systems and MR elastography (MRE) in primary sclerosing cholangitis (PSC). To evaluate the incremental value of combining MRCP+ with morphological scores in associating with biochemical scores. METHODS AND MATERIALS: MRI images, liver stiffness measurements by MRE, and biochemical testing of 65 patients with PSC that were retrospectively enrolled between January 2014 and December 2015 were obtained. MRCP+ was used to post-process MRCP images to obtain quantitative measurements of the bile ducts and biliary tree. Linear regression analysis was used to test the associations. Bootstrapping was used as a validation method. RESULTS: The total number of segmental strictures had the strongest association with Mayo Risk Score (R2 = 0.14), minimum stricture diameter had the highest association with Amsterdam Oxford Prognostic Index (R2 = 0.12), and the percentage of duct nodes with width 0-3 mm had the strongest association with PSC Risk Estimate Tool (R2 = 0.09). The presence of Ducts with medians > 9 mm had the highest association with MRE (R2= 0.21). The strength of association of MRCP+ to Mayo Risk Score was similar to ANALI2 and weaker than MRE (R2 = 0.23, 0.24, 0.38 respectively). MRCP+ enhanced the association of ANALI 2 and MRE with the Mayo Risk Score. CONCLUSIONS: MRCP+ demonstrated a significant association with biochemical scores and MRE. The association of MRCP+ with the biochemical scores was generally comparable to ANALI scores. MRCP+ enhanced the association of ANALI2 and MRE with the Mayo Risk Score. KEY POINTS: • MRCP+ has the potential to act as a risk stratfier in PSC. • MRE outperformed MRCP+ for risk stratifcation. • Combination of MRCP+ with MRE and ANALI scores improved overall performace as risk stratifiers.

Diagnostic interval for non-screening patients undergoing mammography during the COVID-19 pandemic.

OBJECTIVE: During the COVID-19 pandemic, there was a temporary cessation of mammography screening. However, in some facilities, diagnostic breast imaging services continued for patients with a high clinical suspicion of breast cancer. The objective of this study was to evaluate changes in the diagnostic interval (DI) of non-screening patients presenting for diagnostic mammography during the first wave of the COVID-19 pandemic. METHODS: Retrospective chart review was performed on patients presenting for non-screening diagnostic mammography from April 1 to June 30, 2020 (pandemic group) and April 1 to June 30, 2019 (pre-pandemic group). Age, reason for referral, number and type of imaging studies/biopsies necessary for a final diagnosis were recorded. Diagnostic interval (DI) was defined as the number of days from the date of the diagnostic mammogram to the date of the final diagnosis. RESULTS: Compared to the pre-pandemic group (n = 64), the pandemic group (n = 77) showed a reduction in DI of the entire cohort (pandemic: 1 day; pre-pandemic: 15 days, p < 0.0001) for patients not requiring tissue sampling (pandemic: 1 day; pre-pandemic: 11 days, .p < 0.0001) and those requiring tissue sampling with benign pathology (pandemic 9 days; pre-pandemic, 33 days, p = 0.0002). A higher percentage of patients in the pandemic group had their assessment completed during the initial visit (pandemic: 50.6%; pre-pandemic: 23.4%, p = 0.0009). CONCLUSION: During the first wave of the COVID-19 pandemic, the DI for patients with non-screening-related diagnostic mammography was significantly shorter, with a higher percentage of patients completing their assessments on the initial visit, compared to one year prior. KEY POINTS: • Despite reductions in manpower and clinical services, during pandemic times, it is possible to maintain a diagnostic breast imaging service for women at high clinical suspicion for breast cancer. • During pandemic times, breast imaging departments should consider restructuring to a Rapid Diagnostic Unit model with a navigation team that follows patients through the assessment process to a final diagnosis. • Departmental restructuring and patient navigation during pandemic times could either maintain or shorten the diagnostic interval for patients presenting for diagnostic mammography.

ACR Appropriateness Criteria® Chronic Foot Pain.

Chronic foot pain is a frequent clinical complaint, which can significantly impact the quality of live in some individuals. These guidelines define best practices with regards to requisition of imaging studies based on specific clinical scenarios, which have been grouped into different variants. Each variant is accompanied by a brief description of the usefulness, advantages, and limitations of different imaging modalities. The present narrative is the result of an exhaustive assessment of the available literature and a thorough review process by a panel of experts on Musculoskeletal Imaging. The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision include an extensive analysis of current medical literature from peer reviewed journals and the application of well-established methodologies (RAND/UCLA Appropriateness Method and Grading of Recommendations Assessment, Development, and Evaluation or GRADE) to rate the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where evidence is lacking or equivocal, expert opinion may supplement the available evidence to recommend imaging or treatment.

Risk stratification in primary sclerosing cholangitis: comparison of biliary stricture severity on MRCP versus liver stiffness by MR elastography and vibration-controlled transient elastography.

OBJECTIVES: To compare biliary stricture severity on magnetic resonance cholangiopancreatography (MRCP), magnetic resonance elastography (MRE), and vibration-controlled transient elastography (VCTE) liver stiffness (LS) for evaluation of risk stratification and prognostication in primary sclerosing cholangitis (PSC). MATERIALS AND METHODS: Eighty-seven patients (31-61 years; 34 female/53 male) prospectively underwent biochemical testing, VCTE, MRCP, and MRE between January 2014 and July 2016. Correlation between the MRCP grading of PSC based on biliary stricture severity, LS on MRE and VCTE, and the Mayo Risk Score as well as the Amsterdam Oxford Prognostic Index (AOPI) were evaluated and compared. Stricture severity was classified according to previous classification systems based on ERCP. Spearman's correlation and Kruskal-Wallis tests were performed. RESULTS: MRE-LS and intrahepatic stricture severity combined demonstrated higher discriminatory ability among risk categories based on Mayo Risk Score (AUROC = 0.8). MRE-LS alone demonstrated excellent discriminatory ability among risk categories based on AOPI using cutoffs of 1 and 2.7 and was superior to intrahepatic stricture severity (AUROC = 0.9, AUROC = 0.6-0.7). There was a weak correlation between intrahepatic stricture severity and MRE-LS (rho = 0.3; p = 0.011). VCTE-LS values were not correlated with stricture severity and were noncontributory to differentiate patients across risk groups. Intrahepatic stricture severity alone was a poor discriminator of advanced liver fibrosis on MRE (AUROC = 0.7); however, combining intra- and extrahepatic stricture severity and controlling for cholestasis and disease duration improved results (AUROC = 0.9). CONCLUSION: This study demonstrates a significant discriminatory ability of LS values on MRE to distinguish between early to moderate and advanced liver fibrosis. LS values on MRE may add value to risk prognostication and further studies including clinical outcomes are needed. KEY POINTS: • Risk stratification was excellent for liver stiffness measurements on MRE and poor for VCTE and biliary stricture severity. • Risk stratification was further improved when liver stiffness measured on MRE was combined with intrahepatic and extrahepatic stricture severity and indicators of cholestasis were controlled for. • Liver stiffness measurements on MRE correlated with prognostic scores better than measurements performed on VCTE.

Normal Anatomy and Traumatic Injury of the Midtarsal (Chopart) Joint Complex: An Imaging Primer.

The midtarsal (Chopart) joint complex consists of the talonavicular and calcaneocuboid joints and their stabilizing ligaments. Detailed assessment of this complex at MRI can be challenging owing to frequent anatomic variation and the small size of the structures involved. Nevertheless, a wide spectrum of pathologic conditions affect the joint complex, and its imaging evaluation deserves more thorough consideration. This review focuses on MRI evaluation of normal ligamentous anatomy and common variations about the Chopart joint, presenting practical imaging tips and potential diagnostic pitfalls. Imaging findings across a spectrum of traumatic Chopart joint injuries are also reviewed, from midtarsal sprains to Chopart fracture-dislocations. Midtarsal sprains-commonly associated with ankle inversion injuries-are emphasized, along with their often predictable radiographic and MRI injury patterns. Online DICOM image stacks are available for this article. ©RSNA, 2018.

Imaging of Chopart (Midtarsal) Joint Complex: Normal Anatomy and Posttraumatic Findings.

OBJECTIVE: The objective of this article is to review the normal anatomy and posttraumatic findings of the Chopart joint complex. Key imaging features of the normal ligaments and patterns of ligamentous and osseous injuries are discussed. CONCLUSION: Traumatic midtarsal injuries, particularly midtarsal sprain, are often overlooked clinically and on imaging but are relatively common and typically are associated with inversion ankle injuries. Radiologists should be familiar with Chopart joint anatomy and the imaging features of midtarsal injuries because early diagnosis may help optimize clinical management.

Diagnosing Meniscal Pathology and Understanding How to Evaluate a Postoperative Meniscus Based on the Operative Procedure.

Magnetic resonance imaging (MRI) represents the preferred noninvasive imaging technique to diagnose meniscal pathology in the pre- and postoperative setting. Furthermore, characterization of meniscal tissue MR properties has been possible by the development of advanced MRI techniques. Suspected meniscal tears are a frequent indication for MRI and the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine (ISAKOS) classification system has been developed to facilitate accurate and uniform reporting of such meniscal tears. Partial meniscectomy and meniscal suture repair are among the commonly performed procedures and several signs have been described to detect postoperative recurrent tears on MRI. Other techniques that have proven useful for meniscal assessment are ultrasound (US) and computed tomography (CT) arthrography. In recent years, US is being increasingly used in the selective assessment of some meniscal pathology such as tears, parameniscal cysts and meniscal extrusion as it is a relatively inexpensive, accessible, and safe technique. CT arthrography has been advocated as an acceptable alternative in patients with contraindications for MRI, with comparable diagnostic performance.

Ankle impingement syndromes: an imaging review.

Ankle impingement syndromes encompass a broad spectrum of post-traumatic and chronic degenerative changes that present with pain on specific movements about the ankle joint. Both amateur and professional athletes are disproportionately affected by these conditions, and while conservative measures can potentially treat an impingement syndrome, definitive therapy is often alleviated surgically. Imaging (including conventional radiography, ultrasound, CT and MRI) plays an invaluable role in the diagnosis and pre-surgical work-up. An anatomically based classification system is useful in these syndromes, as the aetiology, sites of pathology and preferred treatment methods are similarly based on anatomic locations about the ankle. This review focuses on the anatomic locations, pathophysiology, imaging considerations and brief discussion of therapies for each of the major anatomic ankle impingement syndromes.

MR Imaging of the Midfoot Including Chopart and Lisfranc Joint Complexes.

Following a brief description of the normal anatomy and biomechanics of the midfoot, this article focuses on MR imaging features of common osseous, tendon, and ligament abnormalities that affect the midfoot. Discussion of the anatomy and pathology affecting the Chopart and Lisfranc joint complexes, both of which play important roles in linking the midfoot to the hindfoot and the forefoot respectively, is also included.

Imaging the Glenoid Labrum and Labral Tears.

The shoulder joint is the most unstable articulation in the entire human body. While this certainly introduces vulnerability to injury, it also confers the advantage of broad range of motion. There are many elements that work in combination to offset the inherent instability of the glenohumeral joint, but the glenoid labrum is perhaps related most often. Broadly, clinical unidirectional instability can be subdivided into anterior and posterior instability, which usually raise concern for anteroinferior and posteroinferior labral lesions, respectively. In the special case of superior labral damage, potential dislocation is blocked by structures that include the acromion; hence, while damage elsewhere commonly manifests as clinical instability, damage to the superior labrum is often described by the term microinstability. In this particular case, one of the radiologist's main concerns should be classic superior labral anteroposterior lesions. The glenoid labrum is also subject to a wide range of normal variants that can mimic labral tears. Knowledge of these variants is central to interpreting an imaging study of the labrum because misdiagnosis of labral variants as tears can lead to superfluous surgical procedures and decreased shoulder mobility. This article reviews labral anatomy and normal labral variants, describes their imaging features, and discusses how to discriminate normal variants from labral tears. Specific labral pathologic lesions are described per labral quadrant (anteroinferior, posteroinferior, and superior), and imaging features are described in detail. Online supplemental material is available for this article. ©RSNA, 2016.

High-resolution morphologic and ultrashort time-to-echo quantitative magnetic resonance imaging of the temporomandibular joint.

OBJECTIVE: To implement high-resolution morphologic and quantitative magnetic resonance imaging (MRI) of the temporomandibular joint (TMJ) using ultrashort time-to-echo (UTE) techniques in cadavers and volunteers. METHODS: This study was approved by the institutional review board. TMJs of cadavers and volunteers were imaged on a 3-T MR system. High-resolution morphologic and quantitative sequences using conventional and UTE techniques were performed in cadaveric TMJs. Morphologic and UTE quantitative sequences were performed in asymptomatic and symptomatic volunteers. RESULTS: Morphologic evaluation demonstrated the TMJ structures in open- and closed-mouth position. UTE techniques facilitated the visualization of the disc and fibrocartilage. Quantitative UTE MRI was successfully performed ex vivo and in vivo, reflecting the degree of degeneration. There was a difference in the mean UTE T2* values between asymptomatic and symptomatic volunteers. CONCLUSIONS: MRI evaluation of the TMJ using UTE techniques allows characterization of the internal structure and quantification of the MR properties of the disc. Quantitative UTE MRI can be performed in vivo with short scan times.

High-Resolution Qualitative and Quantitative Magnetic Resonance Evaluation of the Glenoid Labrum.

OBJECTIVE: This study aimed to implement qualitative and quantitative magnetic resonance sequences for the evaluation of labral pathology. METHODS: Six glenoid labra were dissected, and the anterior and posterior portions were divided into normal, mildly degenerated, or severely degenerated groups using gross and magnetic resonance findings. Qualitative evaluation was performed using T1-weighted, proton density-weighted, spoiled gradient echo and ultrashort echo time (UTE) sequences. Quantitative evaluation included T2 and T1rho measurements as well as T1, T2*, and T1rho measurements acquired with UTE techniques. RESULTS: Spoiled gradient echo and UTE sequences best demonstrated labral fiber structure. Degenerated labra had a tendency toward decreased T1 values, increased T2/T2* values, and increased T1rho values. T2* values obtained with the UTE sequence allowed for delineation among normal, mildly degenerated, and severely degenerated groups (P < 0.001). CONCLUSIONS: Quantitative T2* measurements acquired with the UTE technique are useful for distinguishing among normal, mildly degenerated, and severely degenerated labra.

Magnetic resonance imaging of the pulleys of the flexor tendons of the toes at 11.7 T.

OBJECTIVE: We obtained high-resolution 11.7-T MR images of the pulleys of the flexor tendons in cadaveric toe specimens. A detailed understanding of toe pulley anatomy as seen with MR is likely to be of benefit in recognizing disease and the effects of trauma. MATERIAL AND METHODS: Six cadaveric toes were imaged with an 11.7-T small-bore MR imaging system using optimized coils. Two-dimensional dual-echo SE scans were obtained in three planes (40 × 40 × 400-µm(3) voxel size, TE = 7/14 ms, TR = 3,500 ms, fat saturation). Three-dimensional spoiled gradient echo scans were obtained (90-150 µm(3) isotropic voxel size, TE = 6 ms, TR = 25 ms, with and without fat saturation). Specimen orientation was with the long axis of the toe either parallel or perpendicular to B0. RESULTS: All the annular (A) pulleys were demonstrated in the great and lesser toes. The A2 pulley in the great and lesser toes and the A4 pulley in the lesser toes were the most substantial pulleys. The A5 pulley, which has not previously been described in the toes, was demonstrated. The cruciform pulleys were also seen and were smaller and thinner. Three tissue layers were seen, and there was evidence of different fiber directions in annular pulleys producing different magic angle effects. CONCLUSION: Detailed anatomy of the pulley system of the flexor tendons was seen on the 11.7-T MR images showing new features and providing a basis for image interpretation. Similarities and differences between the pulley systems in the toes and the fingers were identified.

Posterior medial meniscus root ligament lesions: MRI classification and associated findings.

OBJECTIVE: The purposes of this study were to determine the prevalence of altered MRI appearances of "posterior medial meniscus root ligament (PMMRL)" lesions, introduce a classification of lesion types, and report associated findings. MATERIALS AND METHODS: We retrospectively reviewed 419 knee MRI studies to identify the presence of PMMRL lesions. Classification was established on the basis of lesions encountered. The medial compartment was assessed for medial meniscal tears in the meniscus proper, medial meniscal extrusion, insertional PMMRL osseous changes, regional synovitis, osteoarthritis, insufficiency fracture, and cruciate ligament abnormality. RESULTS: PMMRL abnormalities occurred in 28.6% (120/419) of the studies: degeneration, 14.3% (60/419) and tear, 14.3% (60/419). Our classification system included degeneration and tearing. Tearing was categorized as partial or complete with delineation of the point of failure as entheseal, midsubstance, or junction to meniscus. Of all tears, 93.3% (56/60) occurred at the meniscal junction. Univariate analysis revealed significant differences between the knees with and without PMMRL lesions in age, medial meniscal tear, medial meniscal extrusion, insertional PMMRL osseous change, regional synovitis, osteoarthritis, insufficiency fracture (p=0.017), and cruciate ligament degeneration (p<0.001). CONCLUSION: PMMRL lesions are commonly detected in symptomatic patients. We have introduced an MRI classification system. PMMRL lesions are significantly associated with age, medial meniscal tears, medial meniscal extrusion, insertional PMMRL osseous change, regional synovitis, osteoarthritis, insufficiency fracture, and cruciate ligament degeneration.

Morphologic characterization of meniscal root ligaments in the human knee with magnetic resonance microscopy at 11.7 and 3 T.

OBJECTIVE: To determine the feasibility of using MR microscopy to characterize the root ligaments of the human knee at both ultra-high-field (11.7 T) and high-field (3 T) strengths. MATERIALS AND METHODS: Seven fresh cadaveric knees were used for this study. Six specimens were imaged at 11.7 T and one specimen at 3 T using isotropic or near-isotropic voxels. Histologic correlation was performed on the posteromedial root ligament of one specimen. Meniscal root ligament shape, signal intensity, and ultrastructure were characterized. RESULTS: High-resolution, high-contrast volumetric images were generated from both MR systems. Meniscal root ligaments were predominantly oval in shape. Increased signal intensity was most evident at the posteromedial and posterolateral root ligaments. On the specimen that underwent histologic preparation, increased signal intensity corresponded to regions of enthesis fibrocartilage. Collagen fascicles were continuous between the menisci and root ligaments. Predominantly horizontal meniscal radial tie fibers continued into the root ligaments as vertical endoligaments. CONCLUSION: MR microscopy can be used to characterize and delineate the distinct ultrastructure of the root ligaments on both ultra-high-field- and high-field-strength MR systems.

Knowledge and power: the asymmetry of interests of Colombian and Rockefeller doctors in the construction of the concept of "jungle yellow fever," 1907-19381.

This study examines the asymmetries among the different interests of officials and medical doctors who worked for the Rockefeller Foundation and their Colombian counterparts in the development and consolidation of the concept of "jungle yellow fever," as distinguished from the known urban form of yellow fever. We explore the research responses to a variety of disease outbreaks in Colombia in the context of the Rockefeller campaigns against yellow fever, from the time of Roberto Franco's initial description of "yellow fever of the forests" in 1907 until the consolidation of the concept of "jungle yellow fever" by Fred Soper in 1938.