Background
Classical machine learning (ML) and deep learning (DL) articles have rapidly captured the attention of the radiology research community and comprise an increasing proportion of articles ...submitted to JMRI, of variable reporting and methodological quality.
Purpose
To identify the most frequent reviewer critiques of classical ML and DL articles submitted to JMRI.
Study Type
Qualitative thematic analysis.
Population
In all, 1396 manuscript journal articles submitted to JMRI for consideration in 2018, with thematic analysis performed of reviewer critiques of 38 artificial intelligence (AI) articles, comprised of 24 ML and 14 DL articles, from January 9, 2018 to June 2, 2018.
Field Strength/Sequence
N/A.
Assessment
After identifying and sampling ML and DL articles, and collecting all reviews, qualitative thematic analysis was performed to identify major and minor themes of reviewer critiques.
Statistical Tests
Descriptive statistics provided of article characteristics, and thematic review of major and minor themes.
Results
Thirty‐eight articles were sampled for thematic review: 24 (63.2%) focused on classical ML and 14 (36.8%) on DL. The overall acceptance rate of classical ML/DL articles was 28.9%, similar to the overall 2017–2019 acceptance rate of 23.1–28.1%. These articles resulted in 72 reviews analyzed, yielding a total 713 critiques that underwent formal thematic analysis consensus encoding. Ten major themes of critiques were identified, with 1‐Lack of Information as the most frequent, comprising 268 (37.6%) of all critiques. Frequent minor themes of critiques concerning ML/DL‐specific recommendations included performing basic clinical statistics such as to ensure similarity of training and test groups (N = 26), emphasizing strong clinical Gold Standards for the basis of training labels (N = 19), and ensuring strong radiological relevance of the topic and task performed (N = 16).
Data Conclusion
Standardized reporting of ML and DL methods could help address nearly one‐third of all reviewer critiques made.
Level of Evidence: 4
Technical Efficacy Stage: 1
J. Magn. Reson. Imaging 2020;52:248–254.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SAZU, SBCE, SBMB, UL, UM, UPUK
The response rate was 100%. The Microsoft 24 Hour Surgery Hololens Group; Jean Florin Ciornohac, CHC Montlégia, Belgium; Amara Gantier Bazan, Foianini Clinic, Bolivia; Bruno Gobbato, Jaragua ...Hospital, Brasil; Bhies Karkazan, Avicenne Hospital, France; Stefan Greiner, Sporthopaedicum, Germany; Ashish Babhulkar, Deenanath Mangeshkar Hospital, India; Michel Ruiz, Angeles Metropolitano Hospital, Mexico; Abderrahim Rafaoui & Belkacem Chagar, CHU Ibn Rochd, Morocco; Stephen Roche, Groote Schuur Hospital, South Africa; Oleksandr Strafun, National Academy of Medical Sciences, Ukraine; Jaber Alkhyeli, Burjeel Medical City, UAE; Roger Emery & Peter Reilly, Imperial College, UK; John Sledge, Lafayette Surgical Specialty Hospital, USA; John Erickson, Atlantic Medical Group, USA; Jeffery C. Wang, University of Southern California Keck Medical School, USA; Jon JP Warner, Harvard Medical School, USA. Local Ethics Committee for the Cochin Hospital Publications located at Site COCHIN, 27, rue du Faubourg Saint-Jacques, 75679 Paris Cedex 14 (email:
On November 13, 2015, terrorist attacks took place in Paris. One hundred and twenty-nine people were immediately killed and 302 needed emergency care. Many resident physicians were on the front line ...of the medical response. Our aim was to report the frequency of symptoms of post-traumatic stress disorder (PTSD), anxiety and depression among resident physicians after the Paris terrorist attacks.
Anonymous questionnaires, including the Impact of Event Scale- Revised (IES-R) and the Hospital Anxiety and Depression Scale (HADS), were emailed two months after the attacks to 2413 Parisian resident physicians. Exposure to the attacks was defined as having direct clinical contact with one of the victims up to one week after the attacks, being one of the victims, or having one among close relatives.
The questionnaire was completed by 680 (28.2%) residents. Eighty-four (12.4%) reported symptoms of PTSD (IES-R ≥ 33), 76 (11.2%) reported symptoms of anxiety (HADS anxiety score > 10) and 16 (2.4%) reported symptoms of depression (HADS depression score > 10). Exposed residents had higher IES-R scores than non-exposed residents (18.8 ± 16.6 versus 14.2 ± 12.0, p = 0.001), and 40 (18.5%) of them reported symptoms of PTSD, compared to 44 (9.5%) of the non-exposed residents (p = 0.001).
There was a high frequency of symptoms of mental distress among our respondents. Dedicated screening and care strategies must be considered in the event of new attacks.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Background
Data about reperfusion injury (RI) following acute arterial mesenteric ischemia (AAMI) in humans are scarce. We aimed to assess the prevalence and risk factors of RI following endovascular ...revascularization of AMI and evaluate its impact on patient outcomes.
Methods
Patients with AAMI who underwent endovascular revascularization (2016–2021) were included in this retrospective cohort. CT performed < 7 days after treatment was reviewed to identify features of RI (bowel wall hypoattenuation, mucosal hyperenhancement). Clinical, laboratory, imaging, and treatments were compared between RI and non-RI patients to identify factors associated with RI. Resection rate and survival were also compared.
Results
Fifty patients (23 men, median 72-yrs IQR 60–77) were included, and 22 were diagnosed with RI (44%) after a median 28 h (22–48). Bowel wall hypoattenuation and mucosal hyperenhancement were found in 95% and 91% of patients with post-interventional RI, respectively. Patients with RI had a greater increase of CRP levels after endovascular treatment (
p
= 0.01). On multivariate analysis, a decreased bowel wall enhancement on baseline CT (HR = 8.2), an embolic cause (HR = 7.4), complete SMA occlusion (HR = 7.0), and higher serum lactate levels (HR = 1.4) were associated with RI. The three-month survival rate was 78%, with no difference between subgroups (
p
= 0.99). However, the resection rate was higher in patients with RI (32% versus 7%;
p
= 0.03).
Conclusion
RI is frequent after endovascular revascularization of AAMI, especially in patients who present with decreased bowel wall enhancement on pre-treatment CT, an embolic cause, and a complete occlusion of the SMA. However, its occurrence does not seem to negatively impact short-term survival.
Key points
Reperfusion injury of the bowel occurred in 44% of patients with acute mesenteric ischemia treated by endovascular revascularization.
A decreased bowel enhancement on initial CT (HR = 8.2), an occlusion of the superior mesenteric artery (HR = 7.0), an embolic cause (HR = 7.4), and higher initial serum lactate level (HR = 1.4) were identified as predictors of reperfusion injury.
Bowel wall hypoattenuation, hyperenhancement of the mucosa, and increase in CRP levels after revascularization were key diagnostic features of reperfusion injury.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
We read with great interest the article by Wanget al.1In their study, the authors applied deep learning tech-niques to predict 1-year risk of nonmelanoma skin cancerfrom clinical diagnostic ...information and medical records,including medication received. Their model showed an areaunder the receiver operating characteristic curve of 0.89(95% CI, 0.87-0.91), which is impressive accuracy.However,the study may have important flaws in its design that arelikely to have biased upward the estimation of the model’saccuracy.First, the authors’ study uses an unmatched case-controldesign. As shown in their Table 1, the 2 groups are very dif-ferent (control patients were a mean of 18 years youngerthan case patients, with half the numbers of medicationsand codes in theInternational Classification of Diseases,Ninth Revision, Clinical Modification). Therefore, it is likelythat one would learn not how to predict a diagnosis of skincancer but simply how to differentiate young people withfew diseases and drugs from older people with several dis-eases and drugs. The authors should have reduced this con-founding bias by, for instance, matching control patientswith case patients according to the known risk factors forskin cancer (eg, age alone yields an area under the curve of0.882).Second, because of the study’s case-control design, the ab-solute risk of 1-year skin cancer was very high because the ra-tio of case patients to control patients was 1:4. The positive pre-dictive values (calledprecisionin the article, following usualmedical informatics terminology) increase with the preva-lence of the disease. In the general population setting, the pre-cision would be much lower than the number reported in thearticle (0.571). For instance, with a sensibility of 0.831 and aspecificity of 0.823, the precision drops to less than 0.05 for aprevalence of 1%.Third, no calibration was reported despite this being es-sential to evaluate prediction models.2This is particularly so-bering because the model is likely to need recalibration to theabsolute risk of the general population for the reasons dis-cussed previously.3Fourth,althoughtheauthors1usedinternalcross-validation,there was no external validation despite its importance for theassessment of performance measures and what to expect if weuse the model in the real world.4In conclusion, the hype over machine learning and deeplearning techniques should not make us forget the key prin-ciples of clinical epidemiology and biostatistics. A solid studydesign still matters.
The fragility index (FI), i.e., theminimum number of best survivors reassigned to the control group required to revert the statistically significant result of a clinical trial to non-significant, is ...a metric to evaluate the robustness of randomized controlled trials (RCTs). We aimed to assess the FI in the field of HCC.
This is a retrospective analysis of phase 2 and 3 RCTs for the treatment of HCC published between 2002 and 2022. We included two-arm studies with 1:1 randomization and significant positive results for a primary time-to-event endpoint for the FI calculation, which involves the iterative addition of a best survivor from the experimental group to the control group, until positive significance (p <0,05, Log-rank test) is lost.
We identified 51 phase 2 and 3 positive RCTs, of which 29 (57%) were eligible for fragility index calculation. After reconstruction of the Kaplan-Meier curves, 25/29 studies remained significant, among which the analysis was performed. The median (interquartile range (IQR)) FI was 5 (2-10) and Fragility Quotient (FQ) was 3% (1%-6%). Ten trials (40%) had a FI of 2 or less. FI was positively correlated to the blind assessment of the primary endpoint (median FI 9 with blind assessment versus 2 without, p = 0.01), the number of reported events in the control arm (RS = 0.45, p = 0.02) and to impact factor (RS = 0.58, p = 0.003).
Several phases 2 and 3 RCTs in HCC have a low fragility index, underlying the limited robustness on the conclusion of their superiority over control treatments. The fragility index might provide an additional tool to assess the robustness of clinical trial data in HCC.
The fragility index is a method to assess robustness of a clinical trial and is defined the minimum number of best survivors reassigned to the control group required to revert the statistically significant result of a clinical trial to non-significant. Among 25 randomised controlled trials in HCC, the median fragility index was 5, and 10 trials among 25 (40%) had a fragility index of 2 or less, indicating an important fragility.
Display omitted
•The fragility index (FI; a method to assess robustness of a trial) is the minimum number of best survivors reassigned to the control group required to revert the statistically significant result of a clinical trial to non-significant.•Among 25 randomised controlled trials in HCC, the median FI was 5, and 10 trials among 25 (40%) had an FI of 2 or less, indicating an important fragility.•FI was correlated with the blind assessment of the primary endpoint, the number of reported events in the control arm, and the impact factor.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Correct positioning of the glenoid component is an important determinant of outcome in shoulder arthroplasty. We describe and assess a new radiological plane of reference for improving the accuracy ...of glenoid preparation prior to component implantation - the Glenoid Vault Outer Cortex (GvOC) plane.
One hundred and five CT scans of normal scapulae were obtained. Forty six females and 59 males aged between 22 and 30 years. The accuracy of the GvOC plane was then compared against the current "gold standard" - the scapular border (SB). Measurements of glenoid inclination, version, rotation, and offset were obtained using both the GvOC and SB planes. These were then compared to actual values.
The mean difference between version obtained using the GvOC plane and the actual value was 1.8° (-2 to 5, SD 1.6) as compared to 6.7° (-2 to 17, SD 4.3) when the SB plane was used, (p < 0.001). The mean difference between estimates of inclination obtained using the GvOC plane and the actual were 1.9° (-4 to 6, SD 1.6) as compared to 11.2° (-4 to 25, SD 6.1) when the SB plane was used, (p < 0.001).
The GvOC plane produced estimates of glenoid version and inclination closer to actual values with lower variance than when the SB plane was used. The GvOC may be a more accurate and reproducible radiological method for surgeons to use when defining glenoid anatomy prior to arthroplasty surgery.
We aimed to identify modifiable and nonmodifiable risk factors for hemoptysis complicating computed tomography (CT)-guided transthoracic needle biopsy.
All procedures performed in our institution ...from November 2013 to May 2015 were reviewed. Hemoptysis was classified as mild if limited to hemoptoic sputum and abundant otherwise. Presence of intra-alveolar hemorrhage on postbiopsy CT images was also evaluated. Patient- and lesion-related variables were considered nonmodifiable, while procedure-related variables were considered modifiable.
A total of 249 procedures were evaluated. Hemoptysis and alveolar hemorrhage occurred in 18% and 58% of procedures, respectively, and were abundant or significant in 8% and 17% of procedures, respectively. Concordance between the occurrence of significant alveolar hemorrhage (grade ≥2) and hemoptysis was poor (κ=0.28; 95% CI 0.16-0.40). In multivariate analysis, female gender (P = 0.008), a longer transpulmonary needle path (P = 0.014), and smaller lesion size (P = 0.044) were independent risk factors for hemoptysis. Transpulmonary needle-path length was the only risk factor for abundant hemoptysis with borderline statistical significance (P = 0.049).
The transpulmonary needle path should be as short as possible to reduce the risk of abundant hemoptysis during CT-guided transthoracic needle biopsy.