Touchstone Benchmark: Are We on the Right Way for Evaluating AI Algorithms for Medical Segmentation?

Autor: Bassi, Pedro R. A. S., Li, Wenxuan, Tang, Yucheng, Isensee, Fabian, Wang, Zifu, Chen, Jieneng, Chou, Yu-Cheng, Kirchhoff, Yannick, Rokuss, Maximilian, Huang, Ziyan, Ye, Jin, He, Junjun, Wald, Tassilo, Ulrich, Constantin, Baumgartner, Michael, Roy, Saikat, Maier-Hein, Klaus H., Jaeger, Paul, Ye, Yiwen, Xie, Yutong, Zhang, Jianpeng, Chen, Ziyang, Xia, Yong, Xing, Zhaohu, Zhu, Lei, Sadegheih, Yousef, Bozorgpour, Afshin, Kumari, Pratibha, Azad, Reza, Merhof, Dorit, Shi, Pengcheng, Ma, Ting, Du, Yuxin, Bai, Fan, Huang, Tiejun, Zhao, Bo, Wang, Haonan, Li, Xiaomeng, Gu, Hanxue, Dong, Haoyu, Yang, Jichen, Mazurowski, Maciej A., Gupta, Saumya, Wu, Linshan, Zhuang, Jiaxin, Chen, Hao, Roth, Holger, Xu, Daguang, Blaschko, Matthew B., Decherchi, Sergio, Cavalli, Andrea, Yuille, Alan L., Zhou, Zongwei

How can we test AI performance? This question seems trivial, but it isn't. Standard benchmarks often have problems such as in-distribution and small-size test sets, oversimplified metrics, unfair comparisons, and short-term outcome pressure. As a con

Externí odkaz: http://arxiv.org/abs/2411.03670

The need to implement FAIR principles in biomolecular simulations

Autor: Amaro, Rommie, Åqvist, Johan, Bahar, Ivet, Battistini, Federica, Bellaiche, Adam, Beltran, Daniel, Biggin, Philip C., Bonomi, Massimiliano, Bowman, Gregory R., Bryce, Richard, Bussi, Giovanni, Carloni, Paolo, Case, David, Cavalli, Andrea, Chang, Chie-En A., Cheatham III, Thomas E., Cheung, Margaret S., Chipot, Cris, Chong, Lillian T., Choudhary, Preeti, Cisneros, Gerardo Andres, Clementi, Cecilia, Collepardo-Guevara, Rosana, Coveney, Peter, Covino, Roberto, Crawford, T. Daniel, Peraro, Matteo Dal, de Groot, Bert, Delemotte, Lucie, De Vivo, Marco, Essex, Jonathan, Fraternali, Franca, Gao, Jiali, Gelpí, Josep Lluís, Gervasio, Francesco Luigi, Gonzalez-Nilo, Fernando Danilo, Grubmüller, Helmut, Guenza, Marina, Guzman, Horacio V., Harris, Sarah, Head-Gordon, Teresa, Hernandez, Rigoberto, Hospital, Adam, Huang, Niu, Huang, Xuhui, Hummer, Gerhard, Iglesias-Fernández, Javier, Jensen, Jan H., Jha, Shantenu, Jiao, Wanting, Jorgensen, William L., Kamerlin, Shina Caroline Lynn, Khalid, Syma, Laughton, Charles, Levitt, Michael, Limongelli, Vittorio, Lindahl, Erik, Lindorff-Larsen, Kresten, Loverde, Sharon, Lundborg, Magnus, Luo, Yun Lyna, Luque, Francisco Javier, Lynch, Charlotte I., MacKerell, Alexander, Magistrato, Alessandra, Marrink, Siewert J., Martin, Hugh, McCammon, J. Andrew, Merz, Kenneth, Moliner, Vicent, Mulholland, Adrian, Murad, Sohail, Naganathan, Athi N., Nangia, Shikha, Noe, Frank, Noy, Agnes, Oláh, Julianna, O'Mara, Megan, Ondrechen, Mary Jo, Onuchic, José N., Onufriev, Alexey, Osuna, Silvia, Panchenko, Anna R., Pantano, Sergio, Parish, Carol, Parrinello, Michele, Perez, Alberto, Perez-Acle, Tomas, Perilla, Juan R., Pettitt, B. Montgomery, Pietropalo, Adriana, Piquemal, Jean-Philip, Poma, Adolfo, Praprotnik, Matej, Ramos, Maria J., Ren, Pengyu, Reuter, Nathalie, Roitberg, Adrian, Rosta, Edina, Rovira, Carme, Roux, Benoit, Röthlisberger, Ursula, Sanbonmatsu, Karissa Y., Schlick, Tamar, Shaytan, Alexey K., Simmerling, Carlos, Smith, Jeremy C., Sugita, Yuji, Świderek, Katarzyna, Taiji, Makoto, Tao, Peng, Tikhonova, Irina G., Tirado-Rives, Julian, Tunón, Inaki, Van Der Kamp, Marc W., Van der Spoel, David, Velankar, Sameer, Voth, Gregory A., Wade, Rebecca, Warshel, Ariel, Welborn, Valerie Vaissier, Wetmore, Stacey, Wong, Chung F., Yang, Lee-Wei, Zacharias, Martin, Orozco, Modesto

This letter illustrates the opinion of the molecular dynamics (MD) community on the need to adopt a new FAIR paradigm for the use of molecular simulations. It highlights the necessity of a collaborative effort to create, establish, and sustain a data

Externí odkaz: http://arxiv.org/abs/2407.16584

Faster ISNet for Background Bias Mitigation on Deep Neural Networks

Autor: Bassi, Pedro R. A. S., Decherchi, Sergio, Cavalli, Andrea

Bias or spurious correlations in image backgrounds can impact neural networks, causing shortcut learning (Clever Hans Effect) and hampering generalization to real-world data. ISNet, a recently introduced architecture, proposed the optimization of Lay

Externí odkaz: http://arxiv.org/abs/2401.08409

Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization

Autor: Bassi, Pedro R. A. S., Dertkigil, Sergio S. J., Cavalli, Andrea

Publikováno v: Nature Communications 15, 291 (2024)

Features in images' backgrounds can spuriously correlate with the images' classes, representing background bias. They can influence the classifier's decisions, causing shortcut learning (Clever Hans effect). The phenomenon generates deep neural netwo

Externí odkaz: http://arxiv.org/abs/2202.00232

A critical view on autoantibodies in lupus nephritis: Concrete knowledge based on evidence

Autor: Bruschi, Maurizio, Angeletti, Andrea, Prunotto, Marco, Meroni, Pier Luigi, Ghiggeri, Gian Marco, Moroni, Gabriella, Sinico, Renato Alberto, Franceschini, Franco, Fredi, Micaela, Vaglio, Augusto, Cavalli, Andrea, Scapozza, Leonardo, Patel, Jigar J., Tan, John C., Lo, Ken C., Cavagna, Lorenzo, Petretto, Andrea, Pratesi, Federico, Migliorini, Paola, Locatelli, Francesco, Pazzola, Giulia, Pesce, Giampaola, Giannese, Domenico, Manfredi, Angelo, Ramirez, Giuseppe A., Esposito, Pasquale, Murdaca, Giuseppe, Negrini, Simone, Bui, Federica, Trezzi, Barbara, Emmi, Giacomo, Cavazzana, Ilaria, Binda, Valentina, Fenaroli, Paride, Pisan, Isabella, Montecucco, Carlomaurizio, Santoro, Domenico, Scolari, Francesco, Mescia, Federica, Volpi, Stefano, Mosca, Marta, Tincani, Angela, Ravelli, Angelo, Murtas, Corrado, Candiano, Giovanni, Caridi, Gianluca, La Porta, Edoardo, Verrina, Enrico

Publikováno v: In Autoimmunity Reviews May 2024 23(5)