Focal Liver Lesion MRI Feature Identification Using Efficientnet and MONAI: A Feasibility Study.

Liver tumors constitute a major part of the global disease burden, often making regular imaging follow-up necessary. Recently, deep learning (DL) has increasingly been applied in this research area. How these methods could facilitate report writing is still a question, which our study aims to address by assessing multiple DL methods using the Medical Open Network for Artificial Intelligence (MONAI) framework, which may provide clinicians with preliminary information about a given liver lesion. For this purpose, we collected 2274 three-dimensional images of lesions, which we cropped from gadoxetate disodium enhanced T1w, native T1w, and T2w magnetic resonance imaging (MRI) scans. After we performed training and validation using 202 and 65 lesions, we selected the best performing model to predict features of lesions from our in-house test dataset containing 112 lesions. The model (EfficientNetB0) predicted 10 features in the test set with an average area under the receiver operating characteristic curve (standard deviation), sensitivity, specificity, negative predictive value, positive predictive value of 0.84 (0.1), 0.78 (0.14), 0.86 (0.08), 0.89 (0.08) and 0.71 (0.17), respectively. These results suggest that AI methods may assist less experienced residents or radiologists in liver MRI reporting of focal liver lesions.

Evolution of liver transplant waiting list in Hungary between 1995 and 2019. / Májvárólista 1995-2019: a hazai adatok nemzetközi összehasonlítása.

Összefoglaló. Bevezetés: A májtranszplantációs program részeként 1995 óta létezik folyamatosan vezetett várólista Magyarországon. Célkituzés: A legfontosabb várólista-paraméterek megállapítása és nemzetközi összehasonlítása. Módszer: A szerzok az 1995. január 1. és 2019. december 31. között elso májátültetés céljából várólistára helyezett betegek adatait elemezték. Eredmények: Összesen 1722 beteget helyeztek várólistára, 1608 felnottet, 114 gyermeket. A férfiak aránya 51,2%, az átlagéletkor 45,6 év. Az évente regisztrált új jelöltek száma 25 év során közel az ötszörösére emelkedett. A listára helyezés leggyakoribb indikációja a víruseredetu cirrhosis volt (n = 451). Ezt követte a cholestaticus (n = 314) és az alkoholos májbetegség (n = 264). Rosszindulatú daganat, 82%-ban hepatocellularis carcinoma miatt 215 beteget regisztráltak. Krónikus betegségekben az átlagos Model for End-Stage Liver Disease pontszám a regisztráláskor 13,5 volt. A 2018. december 31-ig listára helyezettek (n = 1618) 61%-a részesült májátültetésben, 24%-a várakozás közben meghalt, 7%-a a mutétre alkalmatlanná vált. A mutét elotti medián várakozási ido 248 nap volt a krónikus és 2 nap az akut betegek listáján. A transzplantált tumoros betegek (n = 132) szignifikánsan rövidebb ideig vártak mutétre (medián 115,5 nap), mint a többi krónikus beteg (n = 803, medián 282 nap). Az Eurotransplanthoz való csatlakozás utáni idoszakban (2013. július 1. és 2018. december 31. között) a transzplantációs arány növekedett (67%), a várólista-halálozás (meghaltak + mutétre alkalmatlanná váltak) 24%-ra csökkent. Megbeszélés: A várólista folyamatos bovülése hozzájárult a hazai májátültetési program fejlodéséhez. A hazai várólista diagnózis szerinti összetétele a mások által közöltekkel nagyrészt egyezik. A transzplantáltak aránya a nemzetközi átlagnak megfelelo. A várólista-halálozás és a mutét elotti várakozási ido a magyarországinál alacsonyabb donációs aktivitású vagy jelentosen nagyobb várólistával rendelkezo országokéhoz hasonló. Következtetés: Várólista-paramétereink javításához a transzplantációk számának további növelése szükséges. Orv Hetil. 2022; 163(8): 301-311. INTRODUCTION: The Hungarian liver transplant program including waiting list started in 1995. OBJECTIVE: Evaluation of the wait-list parameters and comparing them with those in the literature. METHOD: Data of patients listed for primary liver transplantation between 1995 and 2019 were analyzed. RESULTS: A total of 1722 recipient candidates were registered on the liver transplant waiting list: 1608 adults (51.2% men) with mean age of 45.6 year and 114 patients aged <18 year. Virus-induced cirrhosis was the leading indication of listing (n = 451) and cholestatic liver diseases (n = 314) and alcoholic cirrhosis (n = 264) thereafter. The mean Model for End-Stage Liver Disease score was 13.5 for those with chronic disease. 61% of 1618 patients listed before December 31, 2018 underwent liver transplantation and 31% were removed from the wait-list for death or clinical deterioration. After joining Eurotransplant (period of 01. 07. 2013-31. 12. 2018), the transplant rate was 67%, the waiting list removal due to death/too sick for operation decreased to 24%. The median waiting time till transplantation was 248 days for those on elective and 2 days on acute list. Patients grafted with malignancy (n = 132) waited significantly shorter time than those with chronic non-malignant liver disease (median 115.5 versus 282 days). DISCUSSION: The composition of our waiting list by primary liver disease was similar to that of countries with large burden of hepatitis C. Transplant rate was average, wait-list mortality and waiting time were in line with those observed in low-donation countries or in the case of large volume waiting list. CONCLUSION: Listing of increasing the number of patients contributed to evolution of our liver transplant program. To improve our parameters, increasing transplant activity is warranted. Orv Hetil. 2022; 163(8): 301-311.

Diagnosis of focal liver lesions with deep learning-based multi-channel analysis of hepatocyte-specific contrast-enhanced magnetic resonance imaging.

BACKGROUND: The nature of input data is an essential factor when training neural networks. Research concerning magnetic resonance imaging (MRI)-based diagnosis of liver tumors using deep learning has been rapidly advancing. Still, evidence to support the utilization of multi-dimensional and multi-parametric image data is lacking. Due to higher information content, three-dimensional input should presumably result in higher classification precision. Also, the differentiation between focal liver lesions (FLLs) can only be plausible with simultaneous analysis of multi-sequence MRI images. AIM: To compare diagnostic efficiency of two-dimensional (2D) and three-dimensional (3D)-densely connected convolutional neural networks (DenseNet) for FLLs on multi-sequence MRI. METHODS: We retrospectively collected T2-weighted, gadoxetate disodium-enhanced arterial phase, portal venous phase, and hepatobiliary phase MRI scans from patients with focal nodular hyperplasia (FNH), hepatocellular carcinomas (HCC) or liver metastases (MET). Our search identified 71 FNH, 69 HCC and 76 MET. After volume registration, the same three most representative axial slices from all sequences were combined into four-channel images to train the 2D-DenseNet264 network. Identical bounding boxes were selected on all scans and stacked into 4D volumes to train the 3D-DenseNet264 model. The test set consisted of 10-10-10 tumors. The performance of the models was compared using area under the receiver operating characteristic curve (AUROC), specificity, sensitivity, positive predictive values (PPV), negative predictive values (NPV), and f1 scores. RESULTS: The average AUC value of the 2D model (0.98) was slightly higher than that of the 3D model (0.94). Mean PPV, sensitivity, NPV, specificity and f1 scores (0.94, 0.93, 0.97, 0.97, and 0.93) of the 2D model were also superior to metrics of the 3D model (0.84, 0.83, 0.92, 0.92, and 0.83). The classification metrics of FNH were 0.91, 1.00, 1.00, 0.95, and 0.95 using the 2D and 0.90, 0.90, 0.95, 0.95, and 0.90 using the 3D models. The 2D and 3D networks' performance in the diagnosis of HCC were 1.00, 0.80, 0.91, 1.00, and 0.89 and 0.88, 0.70, 0.86, 0.95, and 0.78, respectively; while the evaluation of MET lesions resulted in 0.91, 1.00, 1.00, 0.95, and 0.95 and 0.75, 0.90, 0.94, 0.85, and 0.82 using the 2D and 3D networks, respectively. CONCLUSION: Both 2D and 3D-DenseNets can differentiate FNH, HCC and MET with good accuracy when trained on hepatocyte-specific contrast-enhanced multi-sequence MRI volumes.

Diagnosis and Management of a De Novo Urothelial Carcinoma in a Kidney Allograft: A Case Report.

INTRODUCTION: Following renal transplantation, the incidence of malignancies is 3-5 times higher than that of healthy individuals. Among other type of cancers, the risk of urological tumors is also elevated. However, only a few cases of de novo transitional cell carcinomas occurring in renal allografts have been reported. CASE REPORT: A 63-year-old tertiary transplanted male patient was urgently hospitalized for a painless macroscopic hematuria. Ultrasonography revealed pyelectasis and a hematoma in the renal pelvis. A percutaneous nephrostomy tube was inserted. An anterograde pyelography was performed later, where a filling defect was still observable in the location of the previously reported hypoechoic mass. Contrast-enhanced ultrasonography showed enhancement of the lesion. An ultrasound-guided percutaneous biopsy was performed. The histologic evaluation revealed a high-grade transitional cell carcinoma. A whole-body staging computed tomography scan did not show signs of metastatic disease. The renal allograft was surgically removed. No disease progression was observed during the 21-month follow-up period. CONCLUSIONS: Painless hematuria and asymptomatic hydronephrosis occurring after kidney transplantation should raise the possibility of urothelial carcinoma in the kidney graft. Contrast-enhanced ultrasound should be considered as a first-line diagnostic modality because it is easily accessible and does not raise concerns about nephrotoxicity or radiation burden.

Cancer Incidence Among Children in New Caledonia, 1994 to 2012.

New Caledonia (NC) is a small French territory in the Pacific Ocean with a relatively young (32% under 20) and multiethnic population. It is divided into 3 districts: Loyalty Island, the North, and the South, each with specific population characteristics. The aim of this study was to describe childhood cancer age-standardized incidence rate (ASR) in NC, compare it with the estimated one in France , and to determine whether residence and ethnicity may have an influence on intrapopulation ASR. All incident cancer cases diagnosed between 1994 and 2012 in children (0 to 14 y) resident in NC for more than 6 months were included. With 162 registered cases, the ASR for all cancers combined was 142.2 (range, 104.9 to 193.3) cases per million children per year and was not different from what was estimated in France (156.6). However, incidence varies according to the place of residence, with a higher ASR in the Loyalty Island district for several types of neoplasms, but not according to the ethnicity. The small migratory flux between this district and the rest of the territory may have led to these results, but, because of the small number of cases, no conclusion can be drawn.

Malignant peripheral nerve sheath tumor of the liver: a case report.

A large, rapidly growing malignant peripheral nerve sheath tumor (MPNST) of the liver in a young female patient, not associated with von Recklinghausen's disease, is presented. Diagnosis was based on detailed immunohistochemical and electromicroscopic examination beside the characteristic H&E picture. As far as we know, this is the first reported, unambiguously proven "de novo" MPNST in the liver. Differential diagnostic problems are discussed and a review of the literature is given.

[The role of imaging methods in the pre- and intra-operative evaluation, and in post-operative follow-up of living donors in liver transplantation in Hungary]. / A képalkotó módszerek szerepe az élô donoros májtranszplantáció donorjelöltjeinek pre- és intraoperatív vizsgálatában, valamint posztoperatív követésében Magyarországon.

The evaluation protocol for liver donors can vary from centre to centre, but the main points are the same. Medical history, physical examination, common laboratory tests and psychosocial evaluation are followed by imaging, and specific haemostasis and viral serology tests. The first imaging examinations have the aim of excluding any disease; conventional chest radiography and abdominal ultrasound are performed. Liver volume, fat content, and vascular and biliary anatomy are then evaluated with contrast-enhanced, multiphase, multidetector row CT/CTA and MR cholangiography. Ultrasound guided liver biopsy, and in some cases digital subtraction angiography, should also be performed. During the first phase of the donor operation, intraoperative investigations are done: cholangiography for the final evaluation of the biliary tree and ultrasound of the hepatic and portal venous system to help draw the resection plane. Donors have regular imaging examinations in the early postoperative period for early detection of complications: mainly US or CT to check the remnant hepatic vascularisation and fluid collections in the operated area, or X-ray for thoracic disorders. It is recommended that regular checkups are performed in the late postoperative period. The paper describes the imaging protocol for donor evaluation applied at our institute at the beginning of our living related liver transplantation programme.

[Imaging techniques for the pre- and postoperative evaluation of recipients with living-donor liver transplantation]. / Elo donoros májátültetés recipienseinek pre- és posztoperatív képalkotó vizsgálatai.

Whole-liver transplantation is a worldwide-accepted method for treatment of end-stage liver disease. As a result of the shortage of cadaveric livers, split-liver transplantation and living-donor liver transplantation are becoming more common. Preoperative imaging of the donors and recipients are indispensable for surgery planning, while postoperative imaging highly contributes to the success of transplantation by follow-up the transplanted organ and early diagnosis of possible sequelae. The authors delineate both pre- and postoperative imaging of recipients of living-donor transplantation, including the role of conventional x-ray, ultrasound, CT and MRI techniques, and also the different possibilities in the field of interventional radiology.